AutoCAD 2013 Help shock – it no longer sucks

Some months ago, I gave Autodesk several damn good (and thoroughly well-deserved) thrashings over its hopelessly inadequate AutoCAD 2013 Help system. When Autodesk’s Dieter Schlaepfer responded and asked for feedback, he sure got it. There are 142 comments on that one post to date, most of them leaving nobody under any illusions about how short of the mark the new system was.

There is now an updated version of the AutoCAD 2013 Help system. It has been an interminably long time coming, a fact made far worse by Autodesk’s stubborn refusal to provide a CHM stopgap (which could have easily been done on the ship date with minimal resources if the will had been there), but at least an update is here now. Is it any good, though?

I’ve seen fit to give the online version of the updated system a few minutes of my time and I have to say that it’s now way, way better than it was before. In a remarkable turnaround from current standard Autodesk practice, it would appear that customer feedback has not just been listened to, but actually acted on. Honest! Search results make sense. Performance is generally way better than I expected from an online system. There are links to useful things like lists of commands. Things like forward/back mouse buttons work as expected. Various things I expected to suck, simply didn’t. Huh? What’s going on here?

It’s not all brilliant. There are occasional unexpected pauses, but not to excess. A Douglas Adams fan (Dieter?) is clearly responsible for The Hitchhiker’s Guide to AutoCAD. It’s fun, but I’m not convinced it’s particularly useful. The layout is confusing and the content has me somewhat baffled. Is DRAWORDER really one of the first things a beginner needs to know about AutoCAD? Or were there 41 things in someone’s list and that one was thrown in there to make the number up to 42?

That aside, this thing is looking good. Unlike the last effort, I don’t think it deserves Total Existence Failure. But does it just look good in comparison with what went before? Have I been so shocked by apparent adequacy that I have missed some glaring problem? You tell me. Please, give it a fair go and report back on what you think.

If you want to try the offline version, here it is. It’s not immediately obvious, but you’ll need to uninstall the old one first before the new one will install, don’t bother. It’s still the horrible old thing. See I was wrong about AutoCAD 2013 Help, it still sucks for what I think about that.

AutoCAD 2013 Service Pack 1 – Now you see it, now you don’t

Last week, Autodesk released Service Pack 1 for AutoCAD 2013, and then removed it a few days later.

Service Pack 1 for AutoCAD 2013 has been temporarily removed due to a newly discovered fatal error. The AutoCAD team is actively working on resolving this and a new service pack will be posted here as soon as it is available.

This is the sort of thing that Beta testing is supposed to prevent, but in this case it obviously didn’t. Somebody in a position of influence at Autodesk needs to investigate whether this is just a freak one-off, or if there is some systemic weakness within the Beta testing program.

One of the supposed benefits of Cloud software is that there are no updates for users to worry about. It’s all taken care of on the vendor’s server and you’re always using the most up to date version. OK, now fast forward five years and imagine how this scenario would have panned out if AutoCAD was a SaaS product.

You come in one morning to find your AutoCAD keeps crashing, or worse, corrupting your files. It keeps doing this for several days until Autodesk is convinced that the problem lies at its end and reverts the changes on its server while it works out how to fix it. In the meantime, there’s nothing you can do to keep your business running except use one of those old-fashioned copies of AutoCAD you have lying around the place. Except your Subscription agreement only allows you to go back 3 releases. Autodesk no longer supports the release you happen to have handy, and won’t allow the software to be activated. You don’t have a 30-day window because you once evaluated that release on your PC and your 30 days are well and truly gone. Oh, and the file format of the drawings you have been using lately has changed several times and the old release won’t open them anyway.

Why would anyone want to put their business in this position?

Edit: A couple of weeks later, Autodesk released AutoCAD 2013 Service Pack 1.1 to resolve the problem.

Olympic Fencing – Mythbusting the Shin v Heidemann Controversy

This is a departure from the usual subject matter of this blog, but one of the advantages of running my own blog is that I can write what I like on it. This post does have a mention of AutoCAD, but it’s so minor and marginal it’s probably not going to interest many of my usual readers.

Introduction

Now the Olympics are over and a video has been made globally available, I’m going to discuss what happened in the Women’s Epee semi-final between South Korea’s Shin A-Lam and Germany’s Britta Heidemann. The image of Shin sitting disconsolate and alone on the piste, in white on a black background, is one of the iconic images of the London 2012 Olympics. It was replayed at the closing ceremony. Besides making for ‘good’ television, it’s one of the human stories that go to make the Games more than just a vehicle to sell junk food. I’m driven to write about this because I’ve seen a huge amount of complete rubbish written on this subject, mostly by people who have absolutely no idea of the subject about which they are ranting so angrily, but also by some fencers who should know better. I have seen my sport unjustly tainted by inaccurate reporting, demonstrably false accusations, defamatory and untrue statements made against honest fencers and officials, and because this is the Internet, a vast amount of illiterate ranting and ugly racism. This includes both ‘joke’ racist trolling and the real thing, neither of which will be tolerated in comments on this blog. I hope that by delaying this post, most of the morons will have moved on and found something else with which to amuse themselves.

It’s difficult to avoid feeling sorry for Shin in her predicament and outrage at what appears to be a terrible injustice. However, I intend to examine what happened step-by-step and analyse it completely dispassionately, explaining the rules and procedures so they can hopefully be understood by non-fencers. Other than Shin and Heidemann, I will not be using the names of anyone involved, in an attempt to make this as clinical as I can. Where there were failings, I intend to clearly point them out. Equally, where there were not failings, I intend to point that out, too, even if it is contrary to popular belief.

I have no bias to declare here; as a British Australian I didn’t really care who won this bout. I do know and respect a coach who knows and respects Shin’s coach, but I’ve never met or seen anybody directly connected to these events. Fencing is a small world and it wouldn’t surprise me to discover that I am connected to most of the Olympic field via only 2 or 3 degrees of separation.

tl;dr?

This is a very long post. If you can’t be bothered reading it all, feel free to skip to the Mythbusting section. However, I’d appreciate it if you took the time to read all the relevant parts before commenting based on just a subset of my observations.

Background

It’s ironic that an epee bout has been such a controversial event, because it’s generally considered the most straightforward of the three fencing weapons to referee, with the simplest rules. The fencers start in the centre of a 14 metre piste, behind en garde lines 4 m apart.  The whole body is target area, and the first person to hit the other is awarded a point. After a point is scored, both fencers are returned to the en garde lines. If both fencers hit each other within 40 – 50 ms (1/25 to 1/20 of a second), they are both awarded a point and are still returned to the en garde lines.

There is an important and relevant exception to this general double-hit, both-get-a-point-and-return-to-the-middle rule, and that is when the scores are equal and the next hit would win the match. Under such circumstances, a double hit simply stops the bout. The scores do not advance and the fencers are restarted, positioned in basically the same locations they occupied before the hit. What do I mean by this? Well, here are the relevant rules (quoted here as an English translation of the canonical rules, which are in French):

t.17.4. When placed on guard during the bout, the distance between the two competitors must be such that, in the position ‘point in line’, the points of the two blades cannot make contact.

t.17.6. If no hit is awarded they are replaced in the position which they occupied when the bout was interrupted.

t.17.8. The competitors may not be replaced on guard, at their correct distance, in such a way as to place behind the rear line of the piste a fencer who was in front of that line when the bout was halted. If he already had one foot behind the rear line, he remains in that position.

t.17.11. The fencers must come on guard correctly and remain completely still until the command ‘Play!’ is given by the Referee.

t.24. When the order ‘Halt!’ is given ground gained is held until a hit has been given. When competitors are replaced on guard, each fencer should retire an equal distance in order to keep fencing distance.

t.25. However, if the bout has been stopped on account of corps à corps, the fencers are replaced on guard in such a position that the competitor who has sustained the corps à corps is at the place which he previously occupied; this also applies if his opponent has subjected him to a flèche attack, even without corps à corps.

t.27. Should a competitor cross the rear limit of the piste completely — i.e. with both feet — a hit will be scored against him.

To explain some of the fencing jargon here, the position ‘point in line’ is where each fencer stands and holds the sword out with a straight arm pointing horizontally at the other fencer. When fencers are started from any position other than the en garde lines, they should be far enough apart that they are both able to do this without the swords crossing. If a fencer feels that the opponent is too close, it is customary to stand point-in-line, at which the referee will expect the opponent to do likewise while the distance is adjusted accordingly. A corps à corps is when the fencers bump into each other, which isn’t a real issue here. A flèche attack is where one fencer takes a ‘running jump’ past the other. This will become relevant later on.

After the preliminaries are over, the fencers start behind their en garde lines, each 2 m from the centre of the piste. Fencing referees always use French at the top levels, so the Austrian referee (selected by computer randomly from a pool of referees from nations not taking part in the bout) calls to the fencers, “en garde” (get on your guard), “pret?” (are you ready?), and if nobody replies with “non” (no), the fencing starts with the referee’s call of “allez” (go/play/fence). If the referee needs to stop the action, she calls “halt”, pronounced in French with a silent ‘h’ and a short ‘a’. It is unfortunate that the words used to start and stop the bout both begin with an “al” sound, and it’s unfortunate that the word to start the action contains two syllables, but that’s what tradition has given us. It could happen that a referee calls en garde, pret, one or both fencers start early and she calls halt! rather than allez.

In this match, the first fencer to score 15 hits will win, or if nobody gets to 15, whoever scores the most hits within the time available. That time is three periods of three minutes, each separated by a one minute break. If scores are equal at the end, a further period of a maximum of one minute is fenced, at which point there will be a definite winner. There are no draws in fencing. There is no let’s-do-it-all-again option. So let’s have a look at how the bout panned out:

First period

YouTube video link

The bout starts normally enough with en garde, pret, allez and the fencers start moving. Around half a second later, there is a small beep that indicates that the clock has been started. Why doesn’t it start at the beginning of the referee’s “allez”? Because at this level, the referee’s duties are separated from that of the timekeeper. At the level I usually fence, the referee holds a remote control and can time the button press to coincide more closely with the spoken call, but at higher levels the referee is relieved of this burden to better concentrate on the fencing and make clearer hand signals to explain decisions to the audience. The down side to this is that there is always a delay between “al-” where the fencers start and “beep” when the clock starts. The delay varies depending on how well the timekeeper can hear the referee given the background noise, how long it takes them to hear the entire word “allez”, determine it’s not “halt” and press the start button. The reflexes of your average timekeeper are not quite as sharp as your average Olympic fencer, but this isn’t usually a significant problem because it’s the same for everybody and most bouts don’t go down to the last second. Fencers should expect every action to be timed from a point slightly after fencing actually starts, and it happened all the time to varying degrees during the London 2012 event.

Although it’s not obvious on the video, exactly one second after the timekeeper presses the button and about 1.5 seconds after the fencing starts, the clock clicks over from 3:00 to 2:59. For this period, the clock will be counting down using its internal resolution: 2:59.90, 2:59.80…2:59.10…2:59.01 – all of these will be displayed as 3:00. So the bout starts with no movement at all apparent on the clock for well over one full second after fencing actually starts. This will happen at the end too, but much more controversially.

During the first period, Heidemann scores 2 hits, then Shin gets one back to leave the score at 2-1. Towards end of the first period, both fencers back off, the referee calls halt and moves to the first one-minute break period with 6 seconds still to play. This is not actually permitted in the rules, but is generally accepted by convention; neither fencer is particularly interested in attacking and nobody complains about it.

Second period

Video link

No hit is scored for one full minute, which is one of the conditions that triggers a non-combativity rule that is designed to encourage fencers to actually fight each other rather than both waiting for the opponent to move; a real problem in epee. When both fencers back off shortly after, the referee calls halt with 1:52 remaining in the second period and moves on directly to the third period. This is done fully in accordance with the rules, so naturally nobody complains about it.

Third period

Video link

At 2:27 Heidemann attacks and Shin counter-attacks, winning a point to leave them at 2-2. Heidemann starts to get more attacking. At 1:53, double-hit, score 3-3. Fencers return to en garde lines in accordance with the rules. 1:33, another double, 4-4. On restart, Shin steps forward over en garde line as en garde is called, but steps back as referee calls pret. Referee ignores this minor infraction and allows play to continue. At 1:13, another double is scored, 5-5. Heidemann has often adjusted her mask during the bout, and at 0:30 waves to attract the attention of the referee who calls halt. Heidemann brushes hair back from her eyes. Fencers are allowed to do this sort of thing within reason. There is a rule that makes it a yellow card offence to have hair obscuring the target area in foil and sabre, or the fencer’s name in epee, and it’s also an offence if hair has to be adjusted during the bout to prevent this particular target/name obscuration from happening. However, there is no rule preventing a fencer from adjusting their hair to move it out of the way of their eyes so they can fence to the best of their ability.

Edit: a referee of greater experience and ability than myself has suggested that Heidemann could well have been given a yellow card for interrupting the bout without adequate justification.

The last 30 seconds pass without a hit. From 0:13, the referee could have called non-combativity which would have resulted in a final minute being fenced in its entirety, regardless of the score unless a fencer reached 15 hits. The referee has some discretion here, because the rule states the trigger for non-combativity as approximately one minute without a valid hit. Instead, she allows play to continue to the end. If she had called non-combativity, it would have removed one potential source of controversy. In a final minute triggered by non-combativity, double hits advance the score and fencers are returned to the en garde lines each time, meaning there is no grey area about where the fencers are supposed to start. This grey area becomes significant later on.

Edit: the same referee has informed me that the approximately one minute period must be continuous, and that the hair adjustment episode, justified or not, constituted an interruption that made the 1:13 period non-continuous.

However, non-combativity isn’t called this time. Therefore, as the last period finished normally with the scores equal, a sudden-death final minute is fenced, with the first single valid hit winning the bout. Double hits stop the bout but are otherwise ignored. But what happens if there is no single valid hit? Who wins then? Do they keep fencing indefinitely until someone wins? No. In situations like this, one fencer is randomly assigned priority at the start of the minute. It could be done with the toss of a coin but is now usually done electronically. What this means is that if the scores are still level at the end of the final minute, the fencer who was awarded priority is given a hit and wins the bout. In this case, priority was awarded to Shin. This means that if the entire final minute had passed without a valid single hit, Shin would have won the bout on what amounted to an electronic coin-toss. Is this fair? Yes. Those are the rules, all fencers enter the event knowing that that is what will happen, and both fencers had an equal chance of being placed in that situation. It is actually very common for a fencer with priority to lose the bout, and extremely rare for the full minute to be fenced. It’s simply a mechanism for forcing the issue and ensuring that there’s at least one fencer with a must-score attitude.

Final minute

Video link

At 0:24 Heidemann performs a flèche attack when Shin has her back foot on the back line, resulting in a double hit. According to rule t.25, Shin is supposed to stay where she is, but instead Shin walks forward and places herself further up the piste. This “stealing” of ground shouldn’t occur, but it’s not usually considered an infringement worthy of penalty unless the referee points it out and the fencer refuses to place herself as requested. The referee fails to do anything about this particular issue and Heidemann does not object. She places herself at the correct distance away from Shin’s new position. When the bout is restarted Shin’s position is such that she has gained 1.65 m. This gives Shin a significant advantage, because if both of her feet move behind the line, a hit is awarded against her and she will lose the bout. This may seem a minor detail to a non-fencer, but it’s not. Being placed on the back line makes it difficult to use distance to time your counter-attacks, a real problem for an epeeist trying to avoid receiving a single hit.

How do I know the distance Shin gained to within about 50 mm? That’s where AutoCAD comes in. I did screen captures of the video footage, paused at the time the hit was scored and the time fencing resumed. In AutoCAD, I attached the images and was then easily able to compare the distance between the fencers’ feet and the back line with known distances (e.g. the back 2 m of the piste). There are probably better tools for the job, but AutoCAD was handy, familiar, and good enough. My job was made easier by the fact that both fencers chose to position themselves on the far side of the piste, right next to the line, making it easy to measure. Strictly, this is against the rules; fencers should position themselves in the centre of the width of the piste. However, referees frequently allow this to happen unless somebody is practically off the side of the piste or a fencer complains. This kind of leeway in minor matters is often shown by referees, as it was in this case throughout the bout.

Back to the bout; at 0:15 Heidemann flèche attacks again, resulting in another double hit. This time, Shin’s back foot is 0.45 m behind the back line when she is hit. The referee checks the video replay to see if Shin went entirely behind the back line, which she did, but not until after the hit, so it didn’t matter. Despite the referee knowing from the replay exactly where Shin should be placed, she allows Shin to walk up the piste again. Before allowing fencing to start, the referee says “distance, both of you” (in French), because the fencers are slightly too close to each other. At this instruction, Heidemann shuffles back slightly but Shin shuffles forwards he same amount and is allowed to start with her back foot is 0.7 m in front of the line. Thus, she gains another 1.15 m.

At 0:09, same again. Heidemann flèches, double hit, Shin moves from 0.7 m behind to 0.65 m in front of the line, a further gain of 1.35 m.

At 0:05, another Heidemann flèche, another double hit. Shin has been pushed back so far that only the front half of her front foot is on the piste and her back foot is 0.6 m behind. That is, she’s about 0.1 m from going over the line completely and losing the bout. She then moves forward so her back foot is 0.4 m in front of the line, a gain of 1.0 m. Before fencing resumes, the referee calls the fencers to ensure adequate distance. Heidemann starts moving forward before pret is called and Shin complains (rightly) about Heidemann being too close, and she moves back slightly. During this distance reset, Shin gains another 0.2 m, giving her a total of 1.2 m advantage this time round. The distance is still slightly too close when fencing resumes, but the referee allows play to continue. Nobody complains about this.

At 0:04, another Heidemann flèche, another double hit, Shin does the forward walk thing again and moves her back foot from 0.5 m behind to 1.1 m in front, gaining another 1.6 m. The distance at resumption is still slightly too close, but nobody complains.

The magic second begins

Video link

After 3 more seconds, at 0:01 another Heidemann flèche results in another double hit. Shin moves her back foot from 0.7 behind to 0.1 in front in preparation for the last second’s play. The distance between the fencers looks too close, so the referee calls for distance again from both fencers. Following this, Shin moves her back foot to 0.1 m behind the line, giving a net gain of 0.6 m this time.

The distance still looks too close based on how close the swords are to each other, but Heidemann is leaning crouched forward which would make the distance look closer than it really is. If both fencers had stood up and presented a point-in-line position to check that the swords could not touch, they may have been not much closer, if any closer at all, than the distance required by the rules. If you doubt this, try it at home with a weapon. Stand en garde facing a wall, then present point-in-line and shuffle forwards until the tip of the weapon touches the wall. Without moving your feet, bend your arm to return to an en garde position. The tip’s not really that far away from the wall, is it? Lean forward as Heidemann did, and you will find that you can easily touch the wall with your weapon, even with a very bent arm.

At 0:01, Heidemann anticipates the call of allez and starts early, with another double hit as the result. If the timekeeper managed to start the clock at all, there is only a tiny fraction of a second (maybe 0.1 s) between the time starting when the timekeeper presses the button and stopping automatically when Heidemann’s point is depressed. As a result, the time still shows 0:01. As a sabre fencer, I assure you that it is not that unusual for an action to be completed within a second; I’ve seen this sort of thing happen before where a fast fencer and a slow timekeeper are involved. The referee could have given Heidemann a yellow card for starting early. However, this would be fairly harsh; a little leeway is usually allowed for this minor infraction, as it was for Shin in the third period. Had a yellow card been awarded, it would have had no immediate direct effect on the bout, but a subsequent infringement of any type would result in a hit against and thus loss of the bout. It would therefore have made Heidemann very wary of starting early again.

When coming en garde this time, Shin resumes in the correct position and Heidemann attempts to start too close. On Shin’s legitimate objection, the referee calls Heidemann to yield distance. She does, but then regains it and a little more. The referee has to call for distance again, and Heidemann is in danger of receiving a yellow card for failing to obey the referee’s instructions. Despite these calls, fencing is allowed to resume with the fencers too close. Again, the situation isn’t as bad as it looks because of Heidemann’s lean forward making the swords look closer than they would be if performing a stand-up point-in-line distance test, but it’s fair to say that the referee did not adequately enforce the rules at this point. This time, it’s to Heidemann’s advantage, unlike the multiple previous times where it was to Shin’s.

At 0:01 again, there is no early start by Heidemann but again, a double hit occurs with a small fraction of a second between the timekeeper’s button-press and the hit landing, and thus the time is still registering 0:01. Shin complains to her coach (apparently about the time not counting down) and walks up the piste again. Some commentators have stated that the timekeeper didn’t start the clock at all during the course of this hit, but I can clearly hear two beeps for the start and the hit, much less than half a second apart. While play is halted, the referee asks the timekeeper, “Time?”, presumably to check that things are working properly. At this, the timekeeper apparently presses the start button accidentally (or mistakenly believing that the referee wished the time to be run down), resulting in it going down to 0:00. It takes about half a second to do this, in hindsight giving some idea of the time that was remaining. Some of the crowd cheered, thinking the match was over. However, the time can’t count down when there is no fencing happening, so there was obviously a fault. Where there is a timekeeping fault, the rules clearly say the referee must estimate the remaining time and have the time set accordingly:

t.32.1. At the expiry of the regulation fencing time, if the clock is linked to the scoring apparatus (obligatory standard for finals of official FIE competitions), it must set off automatically a loud audible signal, and automatically cut off the scoring apparatus, without cancelling hits registered before the disconnection. The bout stops with the audible signal.

t.32.3. Should there be a failure of the clock or an error by the time-keeper, the Referee must himself estimate how much fencing time is left.

She asks for it to be set to one second, which is what the clock said before the error, and calls the fencers en garde. Given that fencing clocks use a visible resolution of one second, this is the only action the referee could reasonably take. It might be that the clock actually had only a hundredth of a second left on it internally, but there’s no way of anyone knowing. She couldn’t estimate the time at 0:00 because there was clearly some time left on the clock when the error occurred; it said 0:01. While the clock is being set back to 0:01, the Korean coach comes out of his technical area to complain about the point that three actions could not be launched without the clock counting down from 0:01. He’s not allowed to come out of his “box”, or appeal any decision (this is the fencer’s responsibility), or shout at the officials. He could well have received a yellow card for doing any of the above. He is apparently a usually very calm, polite gentleman and he must have understandably felt very strongly about the situation to react in this way.

At this point, it’s important to note that it is traditional in fencing (as in most other sports) for all decisions of the referee to be final, and this is largely still the case today. Some decisions can now be appealed, but options are limited. You can appeal on the referee’s interpretation of the rules, but not on a finding of fact by the referee, except in a limited number of video appeals where that facility is available (as it was here). Also, fencing has only one undo step. You can only appeal the previous hit. If you realise that your weapon hasn’t been working all bout when you are 10-0 down, you can ask for it to be tested and you might get the last hit annulled, but 9-0 down is the best you can hope for. As for the time you spent fencing with a useless weapon before that last hit, forget it. That time is gone, and not subject to appeal or adjustment. This may seem harsh and unfair, but it’s really the only way it could be. Trying to unravel a bout beyond a single hit is just too difficult to codify and apply practically, given all of the possible permutations. Here are the relevant rules:

t.122.1. No appeal can be made against the decision of the Referee regarding a point of fact.

t.122.2. If a fencer infringes this principle, casting doubt on the decision of the Referee on a point of fact during the bout, he will be penalised according to the rules. But if the Referee is ignorant of or misunderstands a definite rule, or applies it in a manner contrary to the Rules, an appeal on this matter may be entertained.

t.122.3. This appeal must be made:
a) in individual events, by the fencer,
b) in team events, by the fencer or the team captain,
it should be made courteously but without formality, and should be made verbally to the Referee immediately and before any decision is made regarding a subsequent hit.

Back to the action. At this stage, the fencers are ready to fence the final second. It may be that the time should really have been 0:00.01 rather than 0:01.00, but the latter is the official time at this point. Shin has done a bit of querying and her coach has made his views known, but the coach is not entitled to officially appeal and Shin not done so. Even if an appeal had been made, it is almost certain to have been rejected. The official time is the official time and is not subject to appeal. The referee’s setting of the clock to her estimate of 0:01 was done exactly according to the rules, and is the best estimate that could be made given the one-second resolution that is displayed and can be set on the timing mechanism. An attempt to set a time of some part of a second might possibly have been made by setting to one second and performing a quick start/stop, but it would have been without precedent and fraught with difficulty. An appeal that queried the setting of the time to 0:01.00 would have been rejected on the basis that it related to a matter of fact as determined by the referee. Such an appeal would have possibly resulted in a yellow card for Shin, not that it would have mattered at this late stage. So nothing up to this point has been appealed, and once there’s another hit, nothing that has happened up to this stage can possibly be appealed, regardless of the circumstances.

The referee insists that a reluctant Shin put on her mask and fence the final second that is now on the clock. The referee again calls for distance specifically from Heidemann, although it is Shin who has moved forward during the commotion, gaining another 0.7 m. Again, the fencers are still allowed to start with a distance that is a bit too close, but not as bad as it looks due to Heidemann’s lean. I estimate that the fencers are perhaps 0.5 m too close. Certainly, the distance that the fencers are allowed to start at is inadequate. Equally certainly, the distance is inadequate not because of Heidemann encroaching ahead of where she should, but primarily because of Shin’s walk up the piste.

Up to this point, Shin’s total gain from walking up the piste after each double hit amounts to over 8 m. The referee has erred in allowing this to happen, thereby handing Shin an enormous advantage. It is extremely likely (although not completely certain) that if the referee had set the fencers correctly each time, Shin would have lost by either being driven off the back of the piste, or found herself unable to use backward motion to make effective counter-attacks. This would have happened a long time before the magic everlasting second became an issue. Alternatively, the referee could have warned Shin the first time she did it, issued a yellow card the second time and a red card the third time, handing Heidemann the bout. This could also have happened long before time became a source of controversy. The last second has understandably attracted a lot of attention, but nobody seems to care about the 6 seconds that weren’t fenced in the first period, the 113 seconds that weren’t fenced in the second period, or the 13 seconds that could have been chopped off the third period for non-combativity, or the fact that doing so would have completely changed the situation regarding distance, because each double hit would have returned the fencers to the en garde lines.

The final hit

Video link

So the fencers are set en garde (too close due to Shin’s advancement and the referee’s failure to disallow it) and with a full second on the clock (which is more than there should have been due to the timekeeper’s error). On allez, Heidemann launches forward, beats Shin’s blade and hits her, with the clock still on 0:01. Let’s examine that last sub-second in detail.

First, the referee starts with the words en garde, pret and allez as usual, with the same short period of time between pret and allez that she had been using on previous occasions. This predictability allowed Heidemann to time her start of movement to coincide with the start of the word allez, which she did perfectly. As she launches herself forward, the referee’s word allez is completed, and the timekeeper reacts pretty quickly and presses the start button. You can hear the beep just before Heidemann beats the blade, about 0.3 s after allez starts and only 0.1 s after allez ends. The hit lands shortly after (about 0.7 s later). If the timekeeper was (as some have suggested) deliberately trying to give Heidemann time to score a hit by delaying the start button, they made a really bad job of it. In this particular instance, the timekeeper reacted as quickly as it is reasonable to expect.

The appeals

By the rules of the sport, Heidemann has now won the bout. Heidemann hit Shin one more time than she was herself hit, and this was done within the time allowed as measured on the official equipment. What would normally happen now is that the referee checks the video replay, orders the fencers to the en garde lines, indicates the winner and observes the fencers saluting each other, the referee and the crowd, before shaking hands and retiring to prepare for the next bout. Instead, what happens is that all Hell breaks loose.

This post is already enormous, so I don’t intend to go over the appeal process or the sit-on-the-piste thing in any detail. I will, however, point out that there was really never any hope for any appeal. Remember, nothing prior to the last hit is subject to appeal. Taking that last hit, there was a second left on the official clock at the start, and Heidemann scored a valid hit within that second and before the buzzer loud signalled the end of the bout and the scoring mechanism was disconnected. You can’t undo the fact of that hit just because you don’t like the result, or because it’s heartbreaking for the defeated athlete. However, here are some relevant rules about the appeal process, if you’re interested:

t.84. By the mere fact of entering a fencing competition, the fencers pledge their honour to observe the Rules and the decisions of the officials, to be respectful towards the referees and judges and scrupulously to obey the orders and injunctions of the Referee.

t.95.1. Whatever juridical authority has taken a decision, this decision may be subject to an appeal to a higher juridical authority, but only to one such appeal.

t.95.2. No decision on a question of fact can be the subject of an appeal.

t.95.3. An appeal against a decision only suspends that decision when it can be judged immediately.

t.95.4. Every appeal must be accompanied by the deposit of a guaranty of US$80, or its equivalent in another currency; this sum may be confiscated for the benefit of the FIE if the appeal is rejected on the grounds that it is ‘frivolous’; this decision will be taken by the juridical authority responsible for hearing the appeal. However, appeals against the decisions of the Referee do not require the deposit of the guaranty mentioned above.

Despite the one-appeal rule above, there were actually two appeals entered. The first was an immediate verbal appeal of the referee’s decision (contrary to the rules again, by the coach rather than the fencer), where a gaggle of officials from several different countries spent some time examining the video evidence and discussing the matter. The Technical Director then informed the referee that her decision was upheld, and she awarded the bout to Heidemann who rushed over to shake Shin’s hand as required by the rules, then left to prepare for her next bout. A second, written appeal was then entered, accepted and discussed for even more time while Shin sat alone in tears. The appeal was eventually dismissed, Shin was informed and escorted from the piste by FIE officials.

Mythbusting

Addressing various specific complaints that I’ve seen, let’s take them one at a time to see what’s true and what’s false.

  • Timekeeping in fencing has a resolution of one second, which can lead to problems such as thisconfirmed. But this has always been the case and it’s the same for all fencers. Problems like this are actually extremely rare; when they occur fencers might feel hard done by, but it’s something we accept and live with. That said, I would not be surprised to see a change to fencing timekeeping at the top level as a result of the publicity from this incident, maybe changing to a resolution of 0.1 seconds in the last 10 seconds of the bout.
  • The referee did not place the fencers correctly, allowing Heidemann to start too close on the last two hitsconfirmed. This was indeed a refereeing error; allez should not have been called while they were that close. But it’s important to note that Shin was largely responsible for this because she advanced up the piste and placed herself in that position. It’s also important to note that at no point did either fencer stand point-in-line prior to fencing to ensure the distance was set correctly.
  • Shin advanced up the piste after each double hit but one, gaining vital ground, allowing her to better counter-attack and stay in the boutconfirmed. According to my calculations, she gained over 8 m in this way.
  • The referee demonstrated anti-Shin bias in her placement of the fencersbusted. While it’s true that the referee erred in allowing the fencers to start too close, Shin had placed herself in that position. She could have held out a point-in-line and/or backed off to the correct position in order to ensure there was adequate distance. The idea of anti-Shin bias by the referee is ridiculous, given that she allowed Shin to illicitly gain over half a piste during the final hits.
  • The timekeeper hit the start button a fraction of a second after allez was called, resulting in Heidemann having more time to score the winning hitconfirmed. But this needs to be placed in context. The timekeeper always hits the start button a short time after allez is called, it’s just the degree that varies. The delay that occurred on the winning hit was consistent with what had been occuring earlier in the bout, and indeed in other bouts in the competition. It is not reasonable to expect a timekeeper to react significantly faster than they did on the winning hit.
  • It’s impossible to score two hits without the clock being seen to count down at least a secondbusted. We’re talking about Olympic fencers here, with fast muscles and faster reflexes. The tip of a fencing weapon is said to be the fastest non-ballistic object in sport. Have a look at some of the Olympic sabre bouts and see how much the clock winds down after a few straight attacks from a full 4 m apart. At a much lower level than this, I’ve seen a 5-hit sabre bout decided in 7 seconds, and that’s with the referee doing the timekeeping and thereby avoiding the delays you tend to get with a separate timekeeper.
  • The timekeeper failed to hit the start button at all during the penultimate hitbusted. It is possible to make out a clear start beep during this action, but it’s only just before the beep for the hit.
  • The timekeeper hit the start button while fencing was halted before the last hit, allowing the clock to move from 0:01 to 0:00. This resulted in the clock being reset to 0:01 and Heidemann having a full second to score the winning hitconfirmed. This was indeed an error, and it had a significant effect on the outcome. But it’s no indication of bias. The timekeeper (who I understand to be a British adult fencing volunteer) is unlikely to have had any desire to see any particular fencer win this bout. Fencers generally have an extremely well-developed sense of fair play, so it’s a highly insulting accusation to make. However, if you assume that there was pro-Heidemann bias, doing this deliberately would have been an extremely risky strategy. Not only was the action extremely exposed, there was a significant risk that the time of 0:00 would have been accepted by the referee as the official time, with Shin being declared the winner.
  • The referee erred in having the time set to 0:01 following this timekeeping errorbusted. The rules are explicit about what to do under these circumstances, and the referee followed them in the only way open to her.
  • Heidemann acted ungraciously in celebrating her victorybusted. Nobody who has watched more than a couple of fencing matches could believe this. It’s always unfortunate for the fencer who doesn’t win, but the nature of the sport is such that each bout always has a winner and a loser. Celebrating an important fencing victory with exhuberance is totally normal. Not celebrating a narrow victory to win a place in the Olympic final would have been totally bizarre.
  • Shin received a yellow card during the boutbusted. I have seen several statements to the effect that Shin already had a yellow card and therefore may have felt reluctant to argue with the referee about Heidemann’s distance. This is false. Shin received a yellow card only at the very end of her long wait on the piste, well after the bout was over. What appears to be confusing some observers is that the video shows a patch of yellow next to Shin’s name that appeared between the third period and the final minute. This was there to indicate that she had priority, not to indicate that she had a yellow card.
  • The clock was reset to 0:01 because Shin committed an infringementbusted. I have seen claims that the FIE claimed in an official statement that the clock was reset from 0:00 to 0:01 because of a yellow card infringement by Shin. This is one of the more bizarre things I’ve seen claimed. First, the video makes clear that there was no infringement at that point (at least, not one noticed or addressed by the referee). Even if there had been an infringement, there is no provision in the rules for adjusting the clock because of any infringement. There is no such thing as a “penalty second” in fencing. It didn’t happen, and couldn’t have happened.
  • The clock displays 0:01 when the time is anything from 0:01.99 to 0:00.01 seconds, so there was practically 2 seconds in which to score the last hitbusted. Watch the videos of this and other Olympic bouts and see what the clock does when it’s moving freely (not interrupted by hits). It doesn’t take 2 seconds to move from 0:01 to 0:00. A period starts at 3:00 and moves to 2:59 one second later, not instantly. Moving from 0:02 to 0:01 takes one second. Moving from 0:01 to 0:00 takes one second, too.
  • The clock gets reset to a whole second every time a hit is scoredbusted. Heidemann has been quoted as stating this, but it’s not correct. Again, watch other videos and see what the clock does. Sometimes the clock ticks down practically instantly after the start beep is heard, while on other occasions it takes nearly a full second, depending on how much of a full second was left internally on the clock. The most obvious proof of this is when the timekeeping error is made before the final hit. You can hear the start beep when the button is pressed and the end beep when the time reaches 0:00, and there is clearly less than a full second between the beeps.
  • The FIE reprehensibly demanded money from the Koreans before the appeal, proving that it is corrupt, biased and evil – busted. Some people got pointlessly very worked up about this. This is all above board; it’s written into the rules (see above) and applies to everybody. An $80 deposit is required along with a written appeal. As with other sports, an appeal deposit is there in the established procedures to discourage frivolous appeals. The sum is insignificant in the scheme of things (my last competition’s entry fee was about triple that), and my understanding is that it is almost always refunded anyway. Move along please, nothing to see here.
  • Shin staged a sit-down protest on the pistebusted. Sit-down protests are not allowed in fencing. Any kind of protest outside the official processes (e.g. throwing your equipment around) is generally considered an offence against sportsmanship and subject to a black card. Even refusing to salute correctly is a black card offence, which means exclusion from the whole competition and a 2 month ban. Shin stayed on the piste because she believed, or was told, that she had to remain on the piste while the post-match appeal was being heard. It is true that both fencers staying on the piste is a requirement during an appeal on the piste. However, from my reading of the rules it does not appear to be a requirement during determination of a written appeal.
  • Shin was rudely manhandled off the piste by security goons – busted. Once Shin was informed of the outcome of the official appeal, she had no place on the piste and was under an obligation to leave when requested. There were further bouts to be fenced, including her own bronze medal bout to be held 10 minutes later. If she didn’t leave at this point, she was in grave danger of receiving a black card and losing the right to fight for a medal at all. It was in everybody’s interests, particularly hers, that this didn’t happen. Rather than waving a black card in her face and calling security, a senior FIE official and his colleague put their arms around her and gently encouraged her reluctant departure. A yellow card was discreetly shown at this point; it’s not clear if this was for her initial reluctance to leave or as an automatic procedural result of the failed appeal. It is unfortunate from the point of appearances that the gentlemen in question are significantly larger than Shin, but other than that it is difficult to think how this could have been handled more gently.
  • The FIE should have just made the fencers re-fence the final minute or even the whole boutbusted. Not only is there no provision for doing this within the rules, it’s also an illogical suggestion. People who are outraged about Shin having to hold on for a possible extra second or so are suggesting that she should do so for another whole minute, or up to ten minutes? How does that make sense?
  • Shin “deserved to win”busted. I’m moving away from a purely objective viewpoint by addressing this point, but to me, the fencer who scores the most hits deserves to win. That would be Heidemann, then. If Shin had survived that final everlasting second, she would have been awarded the bout thanks to random selection rather than scoring the most hits. While Shin was undoubtedly disadvantaged by timekeeping errors, she was only still in the bout at that time because Heidemann had already been significantly disadvantaged by refereeing errors.
  • This is the worst example of timekeeping ever seen in top level fencingbusted. Enjoy this video of the second period of the 2009 Veteran (70+) World Championship Sabre Final. Keep an eye on the clock between hits.

Summary

There were undoubtedly mistakes made in both refereeing and timekeeping in this bout. However, they were relatively minor mistakes compared with those that are made on a regular basis by all fencing referees, myself included. It’s not at all uncommon for a hit to be given the wrong way in foil or sabre. It happens, and we live with it, because mistakes are made in all sports where human judgement is involved. Despite a lot of Internet ranting, there is no real evidence of any anti-Shin bias. When examined objectively, the sum of the incorrect and dubious refereeing actions in this bout show that a significant net benefit was provided to Shin.

Although I cannot aspire to the level of excellence and dedication demonstrated by Olympic fencers, I know from recent personal experience that it’s very unpleasant to lose an important bout. It’s heart-rending to lose that bout by the slimmest of margins. It induces not just anguish, but also anger, when you lose that bout because of what you believe to be refereeing, timekeeping or scoring errors. Been there, done that, bought the T-shirt. So I can very much empathise with Shin A-Lam in her predicament. But was she robbed of an Olympic medal? No, not really. Not when you examine all the facts.

AutoCAD 2013 for Mac – the holes live on

A couple of years ago, I reported on the missing features in AutoCAD 2011 for Mac. While some generous souls were prepared to accept something half-baked as a first attempt, even that excuse doesn’t wash when it comes to a third iteration. So how well is Autodesk doing at filling those holes? Decide for yourself. Here’s an updated list of missing features in AutoCAD 2013 for Mac:

  • Quick Properties Palette
  • Layer State Manager
  • New Layer Notification
  • Various layer commands including LAYCUR, LAYDEL, LAYMRG, LAYWALK, and LAYVPI
  • Autocomplete doesn’t work entirely properly, including offering commands that don’t exist
  • Filter
  • Quick Select
  • DesignCenter
  • Tool Palettes
  • Navigation Bar
  • ShowMotion
  • Sheet Set Manager (but there is Project Manager)
  • Model Documentation Tools (but at least now there are object enablers)
  • Geographic Location
  • Table Style Editing
  • Hatch Creation Preview
  • Multiline Style Creation
  • Digitizer Integration
  • Change Space
  • Express Tools
  • Material Creation, Editing, and Mapping
  • Advanced Rendering Settings
  • Camera Creation
  • Walkthroughs, Flybys, and Animations
  • Point Cloud Support
  • DWF Underlays
  • DGN Underlays
  • Autodesk 360 Connectivity
  • Data Links
  • Data Extraction
  • Hyperlinks
  • Markup Set Manager
  • dbConnect Manager
  • eTransmit
  • WMF Import and Export
  • FBX Import and Export
  • Additional Model Import
  • Ribbon Customization
  • Right-click Menus, Keyboard Shortcuts, and Double-Click Customization
  • VisualLISP
  • .NET
  • VBA
  • DCL Dialogs
  • Action Recorder and Action Macros
  • Reference Manager (Standalone Application)
  • Dynamic Block Authoring
  • Custom Dictionaries
  • Password-protected Drawings
  • Digital Signatures
  • Workspaces
  • User Profiles
  • Migration Tools
  • CAD Standards Tools
  • CUI Import and Export

Many of these are big-ticket, dealbreaking items. No DCL? Still? Seriously? To these we can add a whole application, Inventor Fusion, which comes as part of the AutoCAD 2013 for Windows install set. (Edit: Inventor Fusion for Mac is available to download as a Technology Preview application from Autodesk Labs and from the Apple App Store). I don’t expect many Mac users will be heartbroken about the lack of a permanent Ribbon (although there are Ribbon-like things that come and go), but as Autodesk reckons it’s responsible for a 44% productivity boost, maybe they should be. Oh, and it isn’t supported and doesn’t work properly on the current release of OS X.

To be fair, it’s not all one-way traffic. Here’s the list of features that appear only in AutoCAD for Mac:

  • Content Palette
  • Coverflow Navigation
  • Multi-touch Gestures
  • Project Manager (instead of Sheet Set Manager)

Well, that’s all right, then.

We’re used to Autodesk’s unfortunate mastery of the long-term half-baked feature, but carrying on with a whole product this unfinished for three releases is more than a little embarrassing. Charging the same amount for it as real AutoCAD adds insult to injury. While I’m sure there are dozens of Mac users happy to be using anything with the AutoCAD name on it on the platform of their choice, this is not a sustainable state of affairs.

Autodesk really needs to make up its mind about this product before embarking on more Mac misadventures such as porting Revit and other products. Autodesk needs to either take AutoCAD for Mac seriously and finish it off to an acceptable standard, or kill it off as a bad idea. As there’s no sign of the former happening and Autodesk history is replete with examples of the latter, I wouldn’t suggest anybody gets too attached to running AutoCAD natively on OS X.

Source: Autodesk knowledgebase article TS15833488.

AutoCAD Exchange bites the dust

Three years ago, I was happy to promote Autodesk’s then-new site AutoCAD Exchange. However, Autodesk has now given up on this attempt to maintain a social site of its own. My comment at the time, “AutoCAD Exchange is an important and potentially very useful site for AutoCAD users” turned out to be optimistic. I was closer to the mark with “It has yet to be seen if Autodesk manages to develop a real community on this site”. Now we’ve seen the answer. No, it didn’t. Autodesk has instead handed control to more socially successful sites, as this message indicates:

Thank you for visiting AutoCAD Exchange. In an effort to consolidate our online AutoCAD community efforts to more popular networks, we will be migrating AutoCAD Exchange content and activities to Facebook, Twitter and YouTube.

Through the AutoCAD Facebook page we provide tips & tricks, tutorial videos and opportunities to engage with the employees at Autodesk that have helped build the product from the ground up.

We look forward to interacting with you on our social properties.
AutoCAD  on Facebook
AutoCAD on Twitter
AutoCAD on YouTube

Note:AutoCAD Plant Exchange will continue on this site, providing updates, content packs, and productivity tools. Click the Plant Exchange tab or visit www.autodesk.com/plantexchange.

If you are at a corporate location it’s quite likely that Facebook, Twitter and YouTube are strictly off-limits, so I guess your AutoCAD-based social life just fell in a hole. OK, it was a worthy attempt which didn’t work, and it’s understandable that Autodesk pulled the plug. It was by no means the first Autodesk failed venture and it won’t be the last. Companies need to try out new things and nobody succeeds all the time. Apple Newton, anyone? Windows Me? IBM PCjr? AutoCAD for Mac? (The original attempt, that is. Jury is still out on the second one).

What if it had been a more important Autodesk on-line service that didn’t work out? One that you had grown to depend on? When Autodesk gives up on a piece of conventional software you’re using (as has happened countless times in the past, even with successful applications), you can at least go on using it until external factors force you to stop. That might be years down the track. I continued to use Autodesk’s Graphic Impact for presentations when most of Autodesk’s employees had forgotten it ever existed. But when Autodesk gives up on a piece of on-line software you’re using (and you’d have to be mindlessly optimistic to believe that such a thing won’t happen), you’re severely out of luck. So who’s going to volunteer to tie their company in to CAD on the Cloud?

LISP programmers, have your say again

Autodesk wants your input again in its annual API survey. This used to be a closed survey for Autodesk Developer Network (ADN) members, but has been open to all for the last few years. If you do any AutoCAD-based development at all, I encourage you to take part. That includes those of us who do most of our development in LISP.

Here’s the direct link to the survey. As you can see if you click the link, there’s a lot of stuff in there that assumes you’re keen to get developing for AutoCAD WS. If you’re not quite so filled with Cloudy enthusiasm and would prefer Autodesk to expend its resources elsewhere (on fixing and improving Visual LISP, for example), please fill in the short survey and say so. It closes on 22 June, so you only have a week.

Why bother, when it’s obvious that Autodesk is determined to ignore to death its most popular API regardless of whatever anyone says or does? I’m not sure, really. Maybe I’m an eternal optimist. (Ha!) Maybe I just want them to at least feel slightly guilty about sticking their fingers in their ears and going “LALALALALA! NOT LISTENING! WE HAVE A VISION, NOT LISTENING! LALALALALA!”

Cloud concerns – Security – Autodesk puts its arguments on line

I’ve made the point before that while Cloud proponents like Autodesk have been happy to talk big on the potential benefits, they have been conspicuously (suspiciously?) silent on the legitimate concerns their customers have raised. The best responses you have been likely to see regarding such concerns can best be characterised as “glossing over”.

So it’s good to see that Autodesk has put together a white-paper-type-thing called Autodesk® 360: Work Wherever You Are – Safely. This 275 KB PDF, with 5 pages of actual content, puts Autodesk’s point of view about one of the aspects of Cloud that people commonly raise as a concern. This is a good start, but of course there are quite a few potential dealbreakers that need addressing yet.

How well does this document address this issue? As you’d expect from Autodesk, it’s gung-ho positive, but there is at least some acknowledgement of Cloud concerns, e.g. “Customer experiences, however, can be largely impacted by the speed and quality of their Internet connection”. In addition to such occasional connections with the real world, there are some categorical assurances that may make some potential users happier. Here are some examples:

Autodesk 360 is delivered from data centers in the United States.

Files and identities are safe during storage, transit, and usage.

As part of Autodesk‟s due diligence for customer security and protection, prospective Autodesk personnel with potential access to sensitive data are screened through background checks before being employed.

Once purged, your data may persist in backup copies for a reasonable period of time but will not be recovered or read by anyone.

Customers own the content they create.

While that’s all well and good, Autodesk needs to get its legal team reading off the same page and fix up its terms and conditions to make them less anti-customer. Otherwise, this virtual document is worth more less than the virtual paper it’s written on. Some parts of the document are only superficially reassuring:

Any access to customer data or personal information is strictly governed by the Autodesk 360 Terms of Service (http://www.autodesk.com/termsofservice) Autodesk‟s Privacy Policy (http://usa.autodesk.com/privacy/), and internal procedures.

While that sounds fine on the face of it, Autodesk has violated its own Privacy Policy in the recent past and to the best of my knowledge, nothing ever came of it. That assurance is therefore rendered completely worthless by Autodesk’s own history. Add to that the fact that both the Terms of Service and Privacy Policy are extremely rubbery and rather one-sided, and this statement becomes more of a concern than a reassurance.

Other parts of the document are contradictory. Take these statements, for example:

Working in the cloud is all about reducing the hassles and headaches that companies and employees would rather avoid, like …. Managing data security and backups

It’s good practice, however, to personally observe safety and security wherever and whenever you are using Autodesk 360 …. Download and back up work locally in a secure environment.

Cloud’s great because it saves you having to make backups. However, make sure you make backups of the stuff you put on the Cloud. Hmm, okay…

So what do you think of this document? Is it all spin or does it address your security concerns? If it addresses your concerns, does it do so to your satisfaction? Please have a read and add your comments.

Trebling upgrade prices was not enough for Autodesk

A blog post from BIM person Gregory Arkin contains a number of confidently-made statements about what Autodesk intends to do with its upgrade and Subscription pricing model. If the information is correct, the news is all bad for customers. The prices for both upgrade and Subscription are getting jacked up substantially. In fact, for upgraders the pricing (70% of full whack for the cheapest upgrade) will be completely non-viable and you’ll effectively be forced onto Subscription. This goes beyond the trebling of upgrade prices that Autodesk’s Callan Carpenter spent some time defending here two years ago. The link in that post to the relevant Autodesk page doesn’t contain any pricing specifics other than the vague statement “save up to 20%”, but I’ll take Gregory’s word for it.

Gregory sees this business of upgraders being hunted to extinction as something that Subscription customers should have a good laugh about, but he’s wrong. Resellers can have a chuckle, but Subscription customers should mourn the lack of choice for customers. It means customers are no longer able to compare Subscription with any kind of sane upgrade pricing and make a decision about the best option for them. This lack of internal ‘competition’ is not even worth a snigger, because it inevitably means Subscription prices shooting up. Autodesk has racked up Subscription prices already and will do so again next year. For those customers who have fallen for the ‘free’ upgrade-to-suite offers, their Subscription prices will be higher again. With everyone on Subscription, Autodesk will just keep pushing up prices indefinitely. From Autodesk’s point of view, there is no down side to this; tie people in and the gravy train goes on for ever.

There are various terms in common use for this kind of thing. One is price gouging. Anybody who has ever booked a hotel near a big sporting event or bought drinks at a nightclub will be familiar with the realities of businesses doing this whenever they can get away with it. It’s part of the free market and generally perfectly legal, no matter how unpopular it makes the business. However, it only works when competition is effectively absent and customers are left with no realistic alternative but to pay up. Autodesk obviously thinks it is in such a position. Is it right? Sadly, history says it probably is.

Edit: This post originally stated that Gregory is a reseller. He has informed me that this is no longer the case and I have edited the post accordingly. My apologies to Gregory for the misstatement.

Should Autodesk provide a CHM version of its AutoCAD 2013 Help?

I’ve added a poll asking this question over on the right. I would like to see this done as soon as possible as a courtesy for those customers who find the current AutoCAD 2013 Help system inadequate. If you agree, vote Yes. If you disagree (for example, you think Autodesk should instead concentrate on improving the current system), vote No. If you wish to make a comment on this specific issue, feel free.

Autodesk edges towards taking money for Cloud services

In an email to Subscription customers, Autodesk made several announcements about its Autodesk 360 Cloud services.

  1. Subscription users now get 25 GB per seat of Cloud storage, up from 3 GB. Non-Subscription users who create an Autodesk 360 account get 3 GB. The intent here, as with the trebling of upgrade costs, is to get you hooked on Subscription so you become a permanent revenue stream.
  2. More services are now available, apparently, but the list of services looks about the same to me. The table that lists which services are available for which products can be found here. If you’re an AutoCAD user, the only service available is Autodesk 360 Rendering.
  3. The services are now metered. You get a certain number of “cloud units”, and these are eaten up as you use the services. A standard AutoCAD user (with Subscription) gets 100 units. Each render costs you 5 units, so effectively you get 20 on-line renders per seat. That’s enough for a taster, but if Cloud rendering is as brilliant as Autodesk says it is, you’ll soon use that allocation up.
  4. The metering doesn’t mean anything – yet. If you use up all your units, it doesn’t matter. You can go on using more of them as long as you’re on Subscription.
  5. This free lunch will end as soon as Autodesk says so, or as soon as it puts a mechanism in place to charge you for units. No news yet on when that might be, but as parting you from your money is obviously the whole point of the exercise, I can’t imagine it will be too far in the future.

Autodesk reserves the right to change all of this without notice, and to terminate access to Autodesk 360 services at any time and for any reason.

Trusting Autodesk – poll results

I have closed the polls asking if you trusted various companies to do the right thing by their customers. Here is a summary of the results, showing the percentage of “Yes” votes for each company. The most trusted company is at the top, the least trusted is at the bottom.

  1. Honda 69%
  2. Amazon 65%
  3. Target 52%
  4. Bricsys 43%
  5. Apple 36%
  6. Autodesk 23%

Remember, this is not a scientific poll and as with all polls and surveys there will be some self-selection bias. Does anyone find anything about the above results surprising?

ClassicArray Registration Service

When you register ClassicArray, you are supposed to receive an automated response with the registration code. I have not been able to get this working reliably, so for some customers I have been sending out the email manually. Personal circumstances dictate that I will be unable to do this promptly for much of April, so if you purchase ClassicArray and do not receive your code immediately, please accept my apologies in advance.

I will also not be able to be very active on this blog for much of April.

AutoCAD 2013 – An Autodesk Help writer responds

Following some email discussions, I am happy to present a response from a relevant Autodesk person to the posts and comments here about AutoCAD 2013’s Help.

Thanks for the opportunity to respond to your readers, Steve.

First, I want to assure you that we’re listening to your comments about AutoCAD 2013 Help. We are adding it to the valuable feedback we’ve already received from users who participated in our Beta program. I responded promptly to every comment from each Beta user and will now address a wider audience.

Here are some of the trends we’ve seen so far in the suggestions users are making about AutoCAD Help:

  • Additional navigation to supplement our Search function
  • Alphabetical listings of AutoCAD commands and system variables
  • Improved precision from the Search function
  • Access to information about new features

We are working right now on our update plans. We invite your additional comments about any other problems you’ve had so that we can have a broad view of your needs as we define the scope of our update.

I’m glad to say that Help updates will no longer have to wait until the next product release. I’d also like to emphasize that the software industry, Autodesk included, is trending toward online delivery of both software and documentation, and that these technology changes pose significant challenges to all of us.

We always appreciate your feedback and we take it seriously.

Thanks and best regards,

Dieter Schlaepfer
Principal Content Developer
Autodesk, Inc.

As Dieter suggests, please comment here with your problems and suggestions. Please be as brutally honest as you wish about Autodesk, its offerings and its future plans. However, I ask that you remain civil. I have ‘known’ Dieter on-line for about 20 years and can attest to his integrity, intelligence and intense desire to do the best job possible. He did indeed respond very fully and honestly to the comments made by myself and others during the Beta program and in our emails. He may very well be the best listener I’ve ever come across, in any context.

In short, Dieter’s one of the good guys and he’s doing a brave thing here. Let him know what you really think, but please don’t beat him up.

AutoCAD 2013 – Using Help in anger

Trying to be fair, I decided to put aside my initial hostility to the AutoCAD 2013 Help system and use it for real. I used it in a realistic situation, to find out how to work with something new or changed (model documentation) as I was working through it with my own example drawing. Try as I might to give it a fair go, I could only get so far before I got irritated. Using it in anger might not be an entirely appropriate phrase for it, but it’s not that far off. Using it in annoyance, perhaps? Here’s how it went.

I hit F1, wait for it to finish loading itself, click in the search box (because that’s not where the focus is to start with), type ‘model documentation’ and pick Search (because Enter doesn’t work). I then wait again, for about 10 seconds, even though I’ve configured it for offline use. Eventually, there is a huge mass of results displayed, almost all of which are totally (totally!) irrelevant to model documentation. Most of them are relevant only to ARX programmers dealing with completely unrelated matters.

If I use the “phrase” option rather than “and”, the list is much shorter and has a much higher proportion of results that have some relevance, but there are still completely pointless results. For example, the 4th result is About Performance Considerations (AutoLISP), which does not contain the phrase at all. It does contain the words ModelSpace and Document, but not together. It does not contain any information remotely related to model documentation. Didn’t Autodesk buy a search technology company a while back? If that company’s technology is in use here, then Autodesk bought a dud.

At least the top two results directly relate to what I need, so I’ll move on with those. They are Commands for Working With Model Documentation Drawing Views and About Model Documentation. The content of the former page is OK; it’s just a list of commands. There is some pointlessly wasted space at the top of the page that means I have to scroll down to see the bottom of the list, but other than that it serves as a useful reference. The latter page is also fine. It’s an executive summary of the feature with a few relevant pictures, followed by a decent set of links pointing to relevant pages that expand on the subject and explain how to do various tasks associated with it. Now I have overcome the inadequacies of Search and determined that useful Help content is all there, that’s all good then, isn’t it? Not really, I’m afraid.

If Help was being run from a real browser, I’d be able to keep both of those starting pages open with their useful links, then middle-click on each of them as needed to open each useful page in a new tab. However, Help isn’t being run like that. It’s being run from inside Autodesk’s pseudo-browser thing, which only allows one page at a time to be displayed. To be fair, this restriction also applies to the old CHM-based Help to some extent. However, the old CHM Help is split into multiple sections, and it is possible to have multiple CHMs each open in their own windows. For example, I can have the AutoCAD 2010 main Help and Developer Documentation open at the same time, something that’s very important for my productivity and which I would find extremely difficult to give up.

To work around the tabless nature of 2013 Help, I need to choose one particular page and stick to it. When I need another page, I need to navigate back up to one of the original links pages and then back down again. That would be bad enough if navigation within the pseudo-browser was good, but unfortunately it isn’t. Despite what looks like a breadcrumb feature at the top of the page, this is non-functional because of the lack of a hierarchical structure to the content. It just keeps taunting me by saying ‘Home’ and nothing else, pointlessly wasting a swathe of vertical space. There are back and forward buttons in that space, but the back and forward mouse buttons I can use everywhere else do nothing in this browser. You can use Alt+Left and Alt+Right to back and forward. Don’t go too far back, though! If you do, the Search panel goes blank and can’t be restored by going forward again, or by switching between Favorites and Search. To fix this, you can close and restart Help , or pick the Home button and wait about 6 seconds for it to get its act together and restore the Search panel. Then you’ll be at the home page, which may not be where you wanted to go back to.

All right, so I have chosen the single page I’m allowed to have open and I want to use the features it describes. This test PC only has a single 1280 x 1024 screen so there really isn’t room for both AutoCAD and Help at the same time, a situation that will be familiar to users of notebooks. I click on the AutoCAD drawing area behind the Help window, expecting AutoCAD to come to the top and to go behind it. Nothing happens, other than Help losing focus. Help stays on top, obscuring the drawing area. If I click the main AutoCAD taskbar button (this is in XP), that minimizes both Help and AutoCAD. Restoring AutoCAD also restores Help, so it still obscures the drawing area. The two windows are linked, and not in a good way for somebody with one screen. I guess some users will want Help to stay on top, but there are plenty of others who won’t, so what could Autodesk have done to keep everybody happy? Made it configurable, obviously.

Eventually I worked out that I could work on AutoCAD if I explicitly minimised the Help window, so away I went. I used the Commands for Working… page, then the VIEWBASE page to start my model documentation experiment. I then picked another link from the VIEWBASE page, the Drawing View Creation Ribbon Contextual Tab page. Having finished with that, I wanted to get back to the Commands for Working… page, so instead of picking multiple Back buttons (which as noted above is fraught with danger if you do it too often), I clicked on that result in the Search panel on the left. Did this take me back to the Commands for Working… page? No, it did not. It did nothing at all. To make it work I had to click another search result first, and only then the one I really wanted.

One saving grace is that I discovered that if I right-clicked on a link in Help, I could copy the URL and then paste it into a proper browser. This works both on and offline, and allowed me to work around many of the problems noted above. This kludge doesn’t work for search results, though, only for links in pages.

I’ve given AutoCAD 2013 Help a decent go, as much as the average reasonable person would before giving up. Maybe more so. I feel pretty comfortable about giving it what I consider a fair assessment. The content of the Help pages itself looks pretty good to me, at least for those pages I visited and the context in which I was using them. If you already know what command you’re supposed to be using, you just hit F1 from within that command to get at the page you want and you don’t need to go any further, you could well be satisfied. But if you’re using the system in any other way, there’s no getting away from it, it’s a crock. The content is not the problem, it’s the loss of structure to that content, and the browser thing being used to present that content. That loss of structure was A Bad Idea and the browser is a very poor effort. The system as a whole should not have been inflicted on customers.

As a courtesy, Autodesk should do what it did following the 2011 Help debacle and provide a CHM solution for customers to download. It should then go on providing a CHM solution indefinitely, until it can come up with something that is of comparable quality. People are already talking about making their own 2013 CHMs. Autodesk, please do the right thing and save them the bother; let us all know that you’re going to provide CHM as a workaround and get it to us as soon as you can. Don’t worry about losing face by admitting that the 2013 Help isn’t up to scratch. It’s too late for that; we’ve already noticed.

AutoCAD 2013 – Help improved in one area

There’s one important area in which AutoCAD 2013’s Help shines when compared with its immediate predecessors. If you’re a Visual LISP user, you’ll be pleased to know that if you select a function name in the editor (e.g. (vla-get-ActiveDocument)) and hit Ctrl+F1, this now takes you to the appropriate page in the ActiveX and VBA Reference, as it should. In AutoCAD 2011 you just got a cryptic message or a 404 error, depending on the context. In AutoCAD 2012, you were just taken to the front page of Help and expected to find it yourself. Props to Autodesk for fixing this problem.

As a bonus, the reference you’re taken to is still a CHM so it works nicely. The Search tab doesn’t work in Windows 7, but that applies to all CHM Help and it’s Microsoft’s fault, not Autodesk’s. The structured contents and index are fully functional, which makes the whole thing usable even without the search facility.

Autodesk’s Kean about moving to the Cloud

Autodesk’s API guru Kean Walmsley is the second Autodesk person I’ve seen who has been brave enough to stick his head above the parapet by discussing the Cloud, in writing, and in a medium that allows for public comment. Kean has always seemed like a straight shooter to me. Please note that his blog represents his personal opinions rather than an official Autodesk position. He’s after your comments, so please go and let him know what you think on his post. Add your comments here if you’re more comfortable with that, and I’ll make sure Kean sees them.

AutoCAD 2013 – Autodesk video tutorial for Help

Autodesk has produced a 2-minute video explaining the features of the new Help system in AutoCAD 2013 that I recently panned. As you might expect, it’s kind of upbeat and chirpy, but the fact that Autodesk feels the need to provide a tutorial on how to use Help says it all, really. Whatever, it may be useful to you, so here it is. It’s hosted on the Autodesk site, unlike many other Autodesk videos (and my own, to be fair), so those of you who have YouTube blocked at work may still be able to watch it.

If you’re having trouble watching the tutorial, don’t panic. I expect Autodesk will soon produce another tutorial explaining how to use the tutorial explaining how to use Help explaining how to use the product to actually do work.

Autodesk acquires Angry Birds developer Rovio

Autodesk announced today that it had welcomed Rovio Entertainment into the Autodesk fold. Following a US$2.6 billion acquisition, the publisher of mega hit video game Angry Birds is now Autodesk’s Mobile Entertainment division based in Espoo, Finland. “This is a tremendously exciting development for Autodesk going forward,” said Autodesk CEO Carl Bass. “Rovio is the world leader in mobile entertainment software,” he added, “so for Autodesk to have access to that market and that technology opens up a whole new world for us.”

Bass was effusive about the synergistic benefits of the merger and the benefits it will bring to the user interfaces of all Autodesk products. “When a kid starts playing Angry Birds, they don’t need to read a huge mass of documentation. Just show them a few cartoons and they’re away, instantly productive. This is the essence of the democratization of design; it’s not dumbing down, it’s funning up.” This potential ease of use is excellent news for AutoCAD users, because the documentation is now so terrible that it will be wonderful if we no longer have to try to use it.

Former Rovio CEO, Taikke Monniennren, now Autodesk Vice President in charge of Mobile Entertainment, is equally excited about the future. “We have already been given access to the Autodesk code base and my developers can see the potential there. By copying and pasting some code modules, we hope to be able to piece together the Angry Birds 3D Max Suite in a few short years,” said Monniennren. “The main challenge will be in keeping the download size manageable, but with a bit of luck we will be able to keep it down to just a few gigabytes.” In comparison, the original Angry Birds game was just over a megabyte.

“Do not underestimate the strategic importance of this announcement,” said Bass. “Although it may come as an unpleasant surprise to our competitors, our customers are well aware that this is the direction we have been moving for some time now.” This is true. Autodesk is doing whatever it can to appear young, hip, cool, trendy, mobile, social and just totally with it, man. It has been talking and sometimes acting big on Cloud and mobile computing for a while. Clearly, acquiring Visual Tao (now AutoCAD WS) was just the beginning.

Who would dare to call Autodesk antisocial? Autodesk videos are all over YouTube. On Twitter, many key Autodesk people tweet many times a day. The recent AutoCAD 2013 launch in San Francisco was done entirely via Facebook (which enhanced Autodesk’s green credentials by allowing a reduction in the number of press and bloggers flown in from around the world to only 120). The first thing a new AutoCAD 2013 user sees on installation is a Welcome screen that is largely dedicated to Autodesk pushing its app store and Facebook and Twitter pages. Because the Welcome screen phones home on each use, Autodesk can easily slip in new links to any other sites it wishes to promote. I expect your AutoCAD 2013 Welcome screen will sport Angry Birds gaming links within a few days. Angry Birds gaming plug-ins for AutoCAD and related products are likely to appear in the app store soon, but I expect we will have to wait for AutoCAD 2014 until the games become part of the core product.

Clearly for Autodesk, kids are the new adults. But is what spoiled teenagers do on their iPhones really a sound basis for the needs of professional CAD users in a corporate environment? How well does this concept work in practice? Bass answered that by showing a prototype user interface for AutoCAD that uses the new technology. He demonstrated it on a 48″ touch-screen, but it is believed that it will be at least partly functional using old-fashioned mouse-based technology.

Bass started by selecting a red bird from the Ribbon, which sported a snazzy new flouro theme. He dragged the bird down into the drawing area (which had a beautiful animated background with kittens, rainbows and unicorns; let’s hope that makes into production). While this was going on, the system was providing haptic feedback, with the screen vibrating and the bird squawking when dragged close enough to an existing object. By dropping the bird close to the end of a line, Bass was able to start drawing a line from exactly that point. He then drew back the bird and released it such that it was launched in the direction he wanted the line to go. As the bird shot forward, Bass touched it just as it crossed another line and it snapped on to the midpoint with a happy chirp. Perfect!

He then demonstrated other birds in action. To draw a polyline you use the yellow bird and touch it at each vertex as it flings itself along your desired path. To explode a block you use the black bird, triggering a loud explosion which I think will have to be toned down for office use. Ellipse? That white chicken thing that lays an egg. Multileader? The little grey one that splits up when you touch it. And so on. Bass already has his Finnish developers hard at work devising hundreds of new birds that cover most of AutoCAD’s key functionality. Some unimportant features, such as plotting and xrefs, are difficult to translate into birds and will be deprecated into command-line-only versions before being dropped completely in a future release.

The demonstration had to be curtailed after a few minutes when Bass’s arms became too tired for him to continue, but it provided an enticing view of a future where CAD is fun, fun, fun! Addicted users are productive users, according to Bass. “If you’ve ever seen kids playing Angry Birds, you know that they will happily sit there playing it all day every day without complaint. They don’t even stop to eat. The only time they take a break is when they’re forced to visit the bathroom. Even when they’re in there they will probably photograph themselves in the mirror and post it on Facebook. CAD Managers, don’t you wish you could tap that astronomical productivity resource?”

According to Bass, those managers will soon be able to do exactly that. His advice is, “Fire the old fuddy-duddy naysaying Luddites who are allergic to change and replace them with a bunch of kids off the streets. Give them Autodesk software they can’t resist using and they’ll soon be flinging pixels around like there’s no tomorrow. You’ll have an instant office full of the cheapest engineers you’ll ever find, and they’ll be begging you to take their work home with them. With the literally infinite anytime anywhere power of Autodesk 360, they’ll be able to do exactly that. In a few years, the kids in hoodies you see hanging around shopping malls won’t be waiting to snatch your bag, they’ll be leeching wi-fi so they can design your next car on their phones. And they’ll be doing it using Autodesk software.”

Bass refused to be drawn on leaked details about the upcoming iPod Shuffle version of AutoCAD or its supposed marketing slogan Shake to Design, though. “We have a very strict policy of never discussing our plans for future products,” he explained, “except when it suits us.”

The marketing gurus at Autodesk have written an independent productivity report that shows that AutoCAD with the new interface improves productivity by 632.7%. On Windows only, that is. This productivity phenomenon will not apply to AutoCAD for Mac, because there are no plans to provide the Angry Birds interface on OS X. Autodesk believes that this won’t concern Apple users, because Macs are shiny and look really nice.

The impressive productivity figure was generated by performing carefully selected tasks on AutoCAD 2013 using the prototype interface, when using the latest, fastest and most expensive hardware. This was then compared with completely different tasks performed using AutoCAD 1.4. On a twin-floppy IBM PC. With a 12″ monochrome monitor driven by CGA. But without an 8087. The resultant percentage was then multiplied by the number of years since Autodesk produced an AutoCAD feature that wasn’t half-baked on release.

In related news, Autodesk’s legal department has lodged applications to register the words ‘Angry’, ‘Birds’ and all images of feathered flying creatures as Autodesk trademarks. Cease-and-desist notices have already been sent to publishers of ornithologist guide books. Also in Autodesk’s legal sights is Disney Corporation, which clearly violates Autodesk’s intellectual property rights with its depiction of Donald Duck as not just a bird, but frequently as an angry one who goes around smashing things up.

The last word goes to Bass. “Look, the trend is irresistible, and those who can’t keep up will be left behind. Here at Autodesk we believe in freedom of choice. You can either choose to follow our vision of the future, or take to the streets with a cardboard sign and a chipped enamel mug. What could be more democratic than that?”

AutoCAD 2013 – Autodesk pulls off a miracle with Help

In AutoCAD 2011, Autodesk introduced on-line Help. It was badly done and poorly received. It was slow and generally awful to use, and so obviously inferior to the generally well-crafted old CHM-based system in so many ways, that there were squeals of joy when somebody discovered that one of the AutoCAD-based vertical products hadn’t been updated to the new regime and still provided a CHM file. That file became hot property, being posted by users on Autodesk’s own discussion groups and other places. Eventually, the outcry was loud enough that Autodesk was forced to make the CHM version of Help available for download. Those of us who actually use the documentation from time to time (or support people who do) breathed a sigh of relief and got on with our work, grateful that Autodesk had seen the error of its ways. But had it, really?

No. In AutoCAD 2012, Help was not only online, but integrated with AutoCAD Exchange in Autodesk’s dodgy version of a pseudo-browser. How good is Autodesk at writing browsers? About as good as you’d expect, sadly. No AutoCAD 2012 CHM was provided with the product at launch time, or even later as a download.

So how well did this new and improved attempt at on-line Help go down with the punters? In my poll on the worst AutoCAD features of all time, Help (on line / 2012) came in third, which gives you some idea. Third worst of all time! That’s a really, really bad place to be. There’s only one place to go from there, surely?

With AutoCAD 2013, Autodesk has wrought a miracle, taking this terrible failure of a system and completely revamping it. Somehow, incredibly, impossibly, Autodesk has managed to make it even worse. Not slightly worse, either. Much worse. AutoCAD Help has been plucked from the depths of mediocrity and plunged deeper still, to a dark place where the pressure of awfulness is so great that spontaneous implosion is a real risk.

Enough rhetoric, what’s actually wrong with it?

  1. It’s online by default. That means it’s slow and can’t possibly be 100% reliable. This is an example of Autodesk pushing its ‘vision’ at the expense of the practical needs of its customers. Nothing new there, then.
  2. Like many of Autodesk software’s attempts to access the Internet, it has been written poorly, such that it attempts to access IPs directly instead of using calls that will work correctly through a proxy server with firewall. As a result, it fails completely in some secure corporate environments. Mine, for example.
  3. Never mind, let’s just go and tell it to use the offline version instead. Turn off the Options > System > Help and Welcome Screen > Access online content when available toggle, hit F1 and see what happens. This does:
    What? You mean it’s not already installed? I have to download and install it separately myself? To be fair, during installation, you are kind of warned about this with a little information glyph. Hover over it and you will see this:OK, for an individual user that’s a drag, but for a CAD Manager with a large number of computers to set up, that’s way beyond inconvenient. Not to mention a user that’s in a location where online access is expensive, intermittent and/or absent. It would not have been at all hard to include offline Help in a 2013 install, and the fact that a deliberate decision was made to not do so smacks of arrogance. Assuming this worked properly, this situation would reflect pretty badly on Autodesk. But it’s worse than that; read on.
  4. Once the offline Help has been downloaded and installed, let’s try hitting F1 again.  This is what I saw:A little investigative work proved that the Help files were not located in the C:\Program Files\Autodesk\AutoCAD 2013\Help location, but rather in C:\Program Files\Autodesk\AutoCAD 2013 Help\English\Help. That’s right, Autodesk requires you to perform two installations, one of which apparently places files in a different location from that in which the other will go looking. Luckily, instead of going round moving files or changing settings, I just restarted AutoCAD and the problem went away. It’s quite possible I would never have seen the problem if I had done the second installation while AutoCAD wasn’t running, but it’s quite likely that many users will download and install in exactly that situation. The result is confusion, the impression given is amateurish.
  5. When you finally do get to get something other than error messages, this is it:Where’s the User Guide? Command Reference? Customization Guide? Gone. There’s no structure to it. It’s basically just a huge mass of web pages you’re expected to search through. There is no index, no contents, no list of commands, nothing. You want to find Help on something, you go search for it. Not sure exactly what the name is of the feature or command you’re searching for? Bad luck.
  6. You know what Autodesk’s search mechanisms are like, right? Yes, this one sucks too. Beginner? Want to draw a line? In AutoCAD 2010, in the Index tab, start typing LINE and the item appears at the top of the list. Double-click it and you’re there. In AutoCAD 2013, enter LINE in the search box, pick the button and this is what you see:

    So where’s the LINE command? It’s not even on the first page of results. You need to scroll down to find it. No, I’m not kidding. To add insult to injury, unlike every other Windows scroll bar I’ve seen, the one in the search results panel doesn’t page down when you click under the button, it just scrolls a single line at a time. It was probably pretty hard to program it to do that.
  7. Once you’ve got to a page and you want to get around a bit, usability is poor. By default, the search item is highlighted so the title of the LINE page, for example, will display as white on a yellow background. Very readable. There is a general lack of controls, and navigation is awkward and slow. Without a hierarchical structure and no breadcrumb feature, it’s difficult to get an idea where you are and what subjects might also be useful to look at. The tabs that we had in 2010 are gone. There are links to related references and concepts, but they’re at the bottom of each page. If you can see that the page you’re in is not useful to you, you have a bunch of scrolling to before you get to those links. The back and forward buttons on my mouse don’t work here, although they work fine everywhere else.

I could go on and on with the faults, but this post is already a monster so I’ll stop there. It’s difficult to think of ways in which this could have been done worse, but I’m sure Autodesk is working on it for 2014. Seriously, this is just embarrassing.

Autodesk, it’s not as if your software is cheap or sells in tiny quantities. You’re not short of the resources you need to do a good job, yet your AutoCAD Help system is so ridiculously inferior to that of numerous freeware and shareware applications that it’s not funny. What on earth were you thinking when you spewed out this rubbish? Did you think we wouldn’t notice?

You obviously want to convince your customer base that you’re all on-line trendy and capable of providing great Internet-based solutions that will have people flocking to use your Cloud stuff. OK, start by getting your finger out and making this stuff work. After three releases, you can’t do an acceptable job at something so seemingly tailor-made for online use as a bunch of text, images and video (which pretty much describes the World Wide Web). What kind of job can you be trusted to do with something much more difficult, such as CAD on the Cloud?