Tag Archives: Fail

Autodesk remotely killswitches AutoCAD licenses – again

Following the AutoCAD 2019 rollout disaster, where subscription users found their AutoCAD 2018s were broken by an Autodesk licensing system meltdown, Autodesk CEO Andrew Anagnost issued an apology. He also assured customers on Twitter that it wouldn’t happen again:

While I welcomed that, I did have this to say at the time:

I don’t think such a guarantee is realistic, given that the nature of subscription software is to only work when it knows you’ve paid up. At least it demonstrates that the desire is there right at the top to try to prevent such debacles from occurring in future.

Autodesk watchers know that words mean little and actions are everything. I look forward to Andrew sharing news of the actions he’ll be taking to make good on his promise.

Back to the present. Does it come as a surprise to anybody to discover that less than a month later, Autodesk has again accidentally remotely killswitched people’s licenses? Again, not just the new release, but 2018 too. As reported on Twitter and www.asti.com:

After a couple of exchanges, the always-responsive people on the @AutodeskHelp Twitter feed came back with this:

So it’s fixed, but CAD Managers now have a bunch of work to do at their end to deal with a problem created at the Autodesk end. Somebody screwed up, they fixed it, another apology is issued, stuff happens, life goes on.  That’s it for this month, probably. But it doesn’t address a very fundamental problem.

Let’s go back to basics.

The primary function of a licensing system is to allow use of the product by legitimate users.

Everything else is secondary. If a system relies on phoning home and getting the required response from a remote system before you’re permitted to use the software you paid for, not just after installation but on an ongoing basis, then it’s fundamentally flawed from the user’s point of view. Yet that is the system that Autodesk has chosen to base its business on.

No amount of sincere, deeply felt apologies or it’ll-never-happen-again promises will alter the fact that Autodesk has put its own convenience ahead of the ability of paying customers to use what they have paid for. Will Autodesk change that state of affairs? I very much doubt it, in which case it’s inevitable that this kind of thing will happen again, whatever anyone might promise.

AutoCAD 2019 rollout disaster

If you’re an AutoCAD user, you may have been intrigued by the news about the new way Autodesk is bundling up AutoCAD 2019 with various verticals (perpetual license owners need not apply). This is Autodesk’s latest attempt to promote its subscription model and raise prices again. 7% this time, but much more to come. But never mind that, the main point is that you’re getting a whole lot of stuff, and who could say that’s not a good thing?

So it’s most unfortunate for Autodesk that the AutoCAD 2019 rollout has been an unmitigated disaster.

My own experiences in trying to download the product were sub-optimal, but I later learned that I was one of the lucky ones. Subscription users not only had problems with AutoCAD 2019, they found their AutoCAD 2018s were broken too, victims of an Autodesk licensing system meltdown. That’s right, the AutoCAD 2019 launch acted as a remote kill-switch for people’s production software.

Perpetual license users were fine, of course. Only subscription customers suffered. There’s a lesson there that should not go unnoticed.

Back to my own experiences, to begin with I couldn’t download AutoCAD 2019 anywhere, despite it being having been announced everywhere some hours earlier. Autodesk Account wasn’t showing 2019, but it was showing this:

The usual workaround method of downloading the free trial wasn’t any better; that was still on 2018. I even tried installing the execrable Autodesk desktop app (temporarily!), but there was no sign of AutoCAD 2019 there either.

I’ve never had any success using Autodesk’s AVA bot for any real questions, but there was a time when it served as a gateway to Akamai-free downloads, so I tried that. Unfortunately, AVA has been “improved” and that once-excellent feature is gone. I did try asking AVA for the download link to AutoCAD 2019, but that didn’t go well.

First attempt:

Second attempt:

I mentioned my difficulties on Twitter and a few hours later the long-suffering but always-professional @AutodeskHelp people offered this:

Despite appearances, this actually turned out to be helpful because by then the product trial download had been fixed and was offering 2019 rather than 2018. I still didn’t see a direct link and against my better judgement tolerated a very temporary installation of Autodesk’s nasty Akamai download manager. About 5 minutes into the download it said it was going to take about 9 hours so I went away and left it. I returned within the hour to find it was finished, so I have no idea how long it actually took. After immediately uninstalling all traces of Akamai and Autodesk desktop app, I was able to start my evaluation.

So how’s the product? Pretty much the same as 2018, except with a better drawing compare tool. Oh, and the icons are better. The first command I ran in the new release reminded me that some things never change. The ancient rectangular pickbox bug popped in to see me, just like an old friend.

I’ve been warmed by the familiar, comforting experience of this and various other early 90s-era bugs and limitations, where the easiest workaround is to use a competitor’s product to do those bits. More on that in a future post.

An hour or so into using the product saw it crash and burn for the first time:

Don’t take my word for it, though; here are some comments from various users, CAD Managers and long-term Autodesk fans, most of whom had a worse time of it than I did. It’s not just AutoCAD; LT and Navisworks are definitely affected and there may be others. Comments here have been reproduced from various sources under fair use; names have been removed to protect the innocent.

I think it is safe to say that Autodesk are having a very bad day. Account and activation issues, subscription access issues, forum login issues, kudos and comments blocked to some users…

I installed the brand new @Navisworks Manage 2019 last night, and this morning I get the “Your trial has expired” !!!

I can safely say that ‘unimpressed’ is my status right now with the licencing idiocy going on. Not only can I not get the 2019 version, but my 2018 version has stopped working too.

Looks like all subscription (not maintenance or legacy) software is affected this way: logging in and out as well as workarounds posted are not working. Now this is a good reason to abandon Autodesk subscription model.

Good news: Autodesk began rollout of 2019 licenses! Bad news: It might break your 2018 licenses.

Hard to imagine this rollout going any worse.
[Autodesk]”Hold my beer….”

2.5 days without AutoCAD available. I just wish Autodesk would have told us so I could have gone on vacation. Instead, I will have to work thru the weekend to make up lost time.

I love Autodesk and their software, but they have just ditched an entire generation of Autodesk ‘rockstars’ and their subscription portal isn’t working. It really doesn’t look good, does it?

my 2018 is hosed on 3 devices

Unfortunately the situation has lasted for nearly 24 hours and no work done. Can we send the bill to Autodesk?

so I thought I’d try and activate LT
LOL

Although some parts of this are supposedly fixed, it’s not over yet. As I type this, people are still reporting problems and Autodesk is still offering apologies:

Even now, AVA is still completely in the dark about the current release of Autodesk’s traditional flagship product.

It’s great to see users helping each other, though. You may find this Autodesk Forum post by Travis Nave useful as an interim workaround. Also, in a marvellous continuation of McNeel’s free service to the AutoCAD (and BricsCAD) LISP community, DOSLib has been made available for AutoCAD 2019 even before many people were even able to get hold of the software it runs on. That’s what I call service!

All in all, this has not the best start to AutoCAD’s brave new world of oneness. How did it go for you?

That awkward moment when I just failed to create BIM

I recently updated my resume, and I thought it might be relevant to include an episode from my early career. This post is an expansion on what I had to say about that episode.

I was managing a tiny CAD training and development company, Educad. Much of my time there was spent developing software called NIDIS (originally called NEEDS), a project that was started in 1987 or 1988 with Nixdorf Computer as the client. It was intended to take over the market among first the home building companies of Western Australia, then Australia, then the World!

What’s special about NIDIS is that it was a precursor to BIM. Using a 3D-adapted version of the 2D Educad architectural software within AutoCAD, designs of domestic homes could be efficiently created and infused with a degree of intelligence. This was then linked to the Nixdorf minicomputer-based software that contained pricing and other information about the various building components. This combined system enabled accurate quantity take-offs to be performed.

This was supposed to be a short project, but due to a massive amount of “scope creep” it took two years. I was really pushing the limits of what AutoCAD could be persuaded to do at that time and had to break new ground in several areas. Some of it was a kludge, but I made it work. Finally, the software was essentially completed, with a custom tablet menu (remember those?), full documentation and everything. Nixdorf CAD-spec PCs with big screens, tablets, AutoCAD and NIDIS were installed in the drawing office. It was successfully tested in Beta. The take-offs were very accurate. Everything looked good to go.

Then, two weeks before it was due to go into production, this project died. The building company that was sponsoring it, Mansard Homes, went into liquidation as it struggled unsuccessfully with the combined effects of very high interest rates and bad publicity about poor building quality and cost overruns. Nixdorf dropped the whole project like a hot potato and the product was never sold. I didn’t have any rights to the software and couldn’t do anything with it.

But it was software that was based around a 3D model of a building that contained some intelligent information, albeit extremely crude by today’s standards. I wrote, quite literally, Building Information Modeling software. It was completed in 1989, before the name BIM had even been used. The idea had existed since the mid 70s, but I didn’t know that at the time so I made it up as I went along.

I didn’t actually invent BIM, but I made something that resembled BIM that actually worked. And then it didn’t.

This is easily the most spectacular failure of my career. Still, I’m kind of proud of it.

Edit: for historical context, this video shows an unrelated system that was developed at about the same time as NIDIS.

AutoCAD 2018 for Mac – welcome to twenty years ago

In the past, I’ve described how AutoCAD for Mac was released half-baked (as I predicted) and has remained half-baked ever since.

But wait! Autodesk has proudly announced AutoCAD 2018 for Mac. Skimming through that blog post, I must admit my jaw dropped when I saw some of the new features. This one, for example:

This “new feature” was first provided to AutoCAD users in the 20th century. It was an Express Tool in AutoCAD 2000 (released 1999) and was absorbed into mainstream AutoCAD a few years later. The alias editor goes back even further, to the Release 14 Bonus Tools (1997). That one was absorbed into AutoCAD in 1999. Some of the other new features are also old. Migrating your settings was new back in the century that started without powered flight. Now, not so much.

These features are new to AutoCAD for Mac, of course, and that’s kind of the point. Autodesk is advertising, as new, features that were born before some of the adults who are now using their products.

There are other very important features (e.g. DCL support, essential for LISP compatibility) that date back even longer (Release 12, 1992) and which are still missing from AutoCAD for Mac. That’s right, in some areas AutoCAD for Mac is a quarter of a century behind. And counting.

On the bright side, you do now get access to the pointlessly-changed 2018 DWG format. A couple of features are reasonably new additions, but they represent a small subset of the small number of minor improvements in AutoCAD 2017 and 2018 for Windows. If anything, the rate of improvement of AutoCAD for Mac is lagging behind even the glacial progress of AutoCAD for Windows, despite starting from a much lower base point.

I note with interest that Autodesk’s comparison page is now hiding the detail of the differences between the full product and AutoCAD for Mac. I guess if you have two identically-priced products and one’s missing a bunch of stuff, you might be tempted to hide the fact from your potential customers. This post of mine from last year will give you some idea of what Autodesk’s not telling you about what’s missing from AutoCAD for Mac. Clue: it’s a lot.

Mac users pay full price for their product and deserve much better than this. If you want information on a full-featured “AutoCAD for Mac”, don’t bother looking for it from Autodesk. Try Bricsys instead.

The cull continues – yet more Autodesk products are bumped off

While you’re enjoying yourselves at Autodesk University (not that there’s anything wrong with that), spare a thought for a few products that didn’t make it through the year. Their unfortunate ends are unlikely to be announced at AU with flashy videos and gung-ho words, but should still not go unnoticed.

More than just a few products, actually. Autodesk killing off its wares is not new, but 2017 is surely the year where the scythe has been wielded with most gusto. I’ve updated the Autodesk Graveyard again to include a few more ex-products. Thanks to JM and others who have pointed out products that have ceased to be.

While you’re getting excited about subscribing to the latest and greatest new thing, bear in mind that each of the 91 items on the list of demised Autodesk products was once similarly a latest and greatest thing. Also bear in mind that if you’re relying on software that’s cloud-based and/or subscription-only, if the vendor loses interest you could be up a creek without a paddle. You may have to deal with the consequences sooner than you might hope. For example, browser-based renderer Lagoa has been ignoreware since being acquired in 2014 (sound familiar?):

It was only a matter of time, and Lagoa had its pending ending announced on 2 November 2017. It will be put out of its misery on 22 December 2017. That’s not very long for customers to make adjustments.

A reminder: what’s listed on the Autodesk Graveyard is probably incomplete and may not be 100% accurate. Additions and corrections can be made by letting me know in the comments on the post Autodesk products are falling like parrots. If you could provide references that show the birth and death dates of the products you know about, that would be ideal, but all feedback is welcome.

Logitech demonstrates the power of the cloud and cops a bloody nose

I’ve been a pretty satisfied customer of Logitech products for some years. The mice, keyboards, webcams and 3D controllers (branded as 3DConnexion) I’ve used have generally been well designed, well built and long-term software support has usually been very good (with an exception or two). So it’s with some regret that I have to report them as an example of what not to do in customer service.

Logitech recently sent this email to customers of its Harmony Link universal remote control:

This is an important update regarding your Harmony Link. On March 16, 2018,
 
Logitech will discontinue service and support for Harmony Link. Your Harmony Link will no longer function after this date.
 
Although your Harmony Link is no longer under warranty, we are offering you a 35% discount on a new Harmony Hub. Harmony Hub offers app-based remote control features similar to Harmony Link, but with the added benefit of the ability to control many popular connected home devices. To receive your discounted Harmony Hub, go to logitech.com, add Harmony Hub to your cart, and use your personal one-time promotional code […] during checkout.
 
Thank you for being a Logitech customer and we hope you will take advantage of this offer to upgrade to a new Harmony Hub. If you have any questions or concerns about Harmony Link, please email the Harmony customer care team.
 
Regards,
 
Logitech Harmony Team

This isn’t just a matter of no longer supporting an old product (and it’s not that old, anyway – it was still sold directly 2 years ago and old retail stock has been sold until a few months ago). It’s a matter of actively disabling all instances of a product from afar, world-wide.

That’s right, Logitech has demonstrated the (destructive) power of the cloud by using it to remotely kill your perfectly functional device. If it’s out of warranty, send Logitech more money for a newer one. No guarantees on how long it will be before the replacement gets the remote kill-switch treatment.

As you might expect, customers weren’t overjoyed at being treated in this way. Threads popped up on the Logitech forums (where the words “class action lawsuit” were auto-censored), Reddit, Twitter, and as comments on various IT news sites that reported on Logitech’s move.

The supposed reason for Logitech’s decision seemed to make no sense:

We made the business decision to end the support and services of the Harmony Link when the encryption certificate expires in the spring of 2018 – we would be acting irresponsibly by continuing the service knowing its potential/future vulnerability. Our system shows this product, which was last sold by Logitech in fall of 2015, had a small active user base.

Such certificates are commonly purchased and renewed by hardware and software companies for relatively tiny amounts of money. It would have cost Logitech less to renew a certificate than it would to have someone write the explanation about why they weren’t doing it. Very odd. As a business decision, it sucks. It also exposes Logitech management as utterly out of touch with the reality in which their customers live.

In a reaction that should have come as a surprise to nobody (but apparently did to Logitech), pretty much everybody gave the company a major roasting. Many people pointed out that such a move would be considered illegal in their countries (including mine), or at best (for Logitech) it would entitle the customer to a full refund from the retailer. Many people promised to never buy anything from Logitech in particular, and any device capable of being remote-bricked in general.

Once it became apparent that this was a major PR disaster, Logitech did a belated partial U-turn and extended the offer of a free replacement to customers with units that were out of warranty.

“I made a mistake,” head of Logitech Harmony Rory Dooley explains to Wired. “Mea culpa. We’re going to do right by our customers, and do the right thing.”

This reminds me of those politicians who get caught out misusing expenses who then say sorry and offer to pay back the ill-gotten gains, as if that’s enough to get them off the hook. Nope. Too little, too late.

Logitech, you just destroyed a whole bunch of customer trust. How valuable is that to you? How much is it going to cost you in sales? How much will it cost you in marketing to try to regain it? It stands to be a fair bit more than the cost of updating a certificate, I would guess. And you’re still bricking a whole bunch of perfectly functional devices. How is that environmentally responsible?

The idea of any product that can be remote-disabled or even reduced in functionality by anybody should be anathema to all of us. Any product. Not just gadgets. (How’s your internet-reliant juicer going? Oops.) Oven, garage door, fridge, car (Tesla can do this), hardware, firmware, software. Yes, software.

This, ladies and gentlemen, is why we don’t do CAD in the cloud. I’ve explained years ago how cloudy CAD adds multiple points of failure. I’m still not wrong about that. One of those additional points of failure is when the vendor decides to stop offering the service. And, of course, the same applies to subscription software, even when it’s not cloudy. The vendor loses interest and you’re left high and dry.

Don’t think it won’t happen. It happens repeatedly and will continue to happen. Don’t be a victim when it does.

I have closed the three satisfaction rating polls I started a couple of months ago and will be reporting the results individually. The usual caveats about online polls apply.

This poll indicates that Autodesk customer satisfaction levels are perhaps not quite as elevated as they could be. The mean rating is 2.75, the median is 2 and the mode is 0. Yes, zero. Very dissatisfied customers outnumber very satisfied ones by nearly ten to one. That’s pretty emphatic.

If Autodesk had left its policies alone and tried to run a huge PR campaign to deliberately make itself as unpopular with its customers as possible, I doubt that it could have achieved anything like this poor a result. Congratulations, I guess.

Although this is an appalling result for Autodesk, it should come as no surprise to anyone. This reflects the sentiment I see pretty much everywhere, in a wide variety of online forums and when talking to all sorts of people in person.

Lesson for the day: there’s no point in spending a billion dollars a year on marketing if you’re going to do things that make you about as popular as a fart in an elevator.

Autodesk products are falling like parrots

Autodesk is killing off products at such a rate I can’t keep up with it all. The latest ones to fall off the perch and join the choir invisible are Structural Detailing and Advance Concrete.

I think. As I said, I can’t keep up.

Despite the recent departures, Autodesk still has way too many products and it’s inevitable that the cull of Carl’s acquisitions and creations will continue. It’s just too bad if you’re one of the people using a product that Autodesk feels isn’t profitable and/or exciting enough, you’ll just have to learn to live without it.

Although 2017 has been particularly brutal for End Of Life experiences, Autodesk killing off products is of course nothing new. Autodesk is even named after a dead product (well, stillborn).

Trace back through Autodesk’s history and you’ll see a long and bloody trail of product corpses and wailing orphans. Maybe you’re one of those orphans. Tell us all about your experiences if you think it might be cathartic. Were you looked after?

What sort of product does Autodesk kill off? Let’s narrow it down.

Autodesk kills new products, old products, cloud products, desktop products, mobile products, free products, paid products, full products, add-on products, large products, small products, products that were bought just to kill off for anti-competitive reasons, products that have been ignored to death, and products that Autodesk hyped to the heavens as the best thing ever and if you didn’t get on board you’d be left behind with all the other Luddites and look silly and old-fashioned as your competitors strode off arm-in-arm with Autodesk into a wonderful bright future.

You name it, Autodesk has killed it. It would seem that almost nothing is safe.

To give you some idea what I mean, I’ve resurrected my Autodesk Graveyard page. I tried this a few years ago but it was too big a job to create and maintain it so I killed it (ironically enough). Thanks to Edwin Prakaso on Twitter for inspiring me to have another go at it, with a bit less detail this time so hopefully it’s manageable.

Image of war graves by Arne Hückelheim.
No disrespect intended to those who gave the ultimate sacrifice. Lest we forget.

The Autodesk Graveyard is by no means complete and what’s there may not be 100% accurate. Additions and corrections can be made by letting me know in the comments on this post. If you could provide references that show the birth and death dates of the products you know about, that would be ideal, but all feedback is welcome.

Hot tip for Autodesk

Hey Autodesk high-ups, I’m sorry you’ve been having so much trouble persuading your customers to throw away their perpetual licenses and throw themselves on your perpetual mercy. It’s clearly difficult to persuade technical types to do dumb things like rent your software at enormous and ever-increasing prices. I feel for you. But there’s an answer.

Find dumber customers.

Lots of them. And fast, before the stock market notices that you’re no Adobe and we’re not buying it. Sorry, I mean not renting it.

Look no further! Simply buy this company, discard the product when you’re bored with it (you’re very familiar with that process) and get hold of the customer list.

Sell subscription software to those people. They’ll have no idea what they’re renting or why, but that doesn’t matter. They’ll buy anything that’s pretty, hip, now, connected, and preferably organic. They will commit to perpetually shelling out large sums just to keep using it, no matter how poorly it performs. They’re rich and dumber than rocks. All of this makes them ideal customers for you.

If you’re a bit strapped for cash at the moment, just have a word with the investors (including Google) who pumped $120M into an Internet-enabled $700 (sorry, now $400) machine that squeezes expensive pre-squeezed juice out of DRM-protected short-lifespan bags, and manages to do it slower and noisier than you can do it with your bare hands. They’re even dumber than the customers, so squeezing money out of them will be easier than squeezing juice out of a bag when the Wi-Fi’s down.

This is a perfect fit for you, Autodesk. It has everything you need to ensure mission-friendly proactive synergistic compatibility on a going-forward basis. It’s disruptive. It looks good. It’s an overpriced, poorly functioning product. It has on-point (but pointless) compulsory connectivity. It ties users into paying whatever you ask, for ever. And best of all, it connects you to a collection of completely clueless cashed-up customers.

Thanks to @internetofshit on Twitter for making me aware of this and other hilarious Internet of Things (IoT) idiocy. Examples:

Enjoy!

The cloud broke
And teardrops fell
On the desks
Of those who fell
For the lure
Of a cloudy hell.


The landlord laughs
To see such fun
Collects his rent
From web he spun
He still gets paid
When things don’t run.


I said t’would be
It’s come to pass
Surprised? Not me
With or without Bass
Autodesk’s cloud
Can kiss this SaaS.

Disaster in progress – Getting it wrong

No, not Autodesk getting it wrong, me  getting it wrong. In recent posts, I supported my arguments against Autodesk’s move to all-rental software with faulty evidence. As pointed out to me by several commenters, I completely failed to take deferred revenue into account. I would like to sincerely thank those who pointed out my error.* Although I included a disclaimer about not being a financial analyst, I should have gone further and simply not ventured into areas I am ill-qualified to cover. I got it wrong. I therefore offer unreserved apologies to Autodesk and my readers.

What now?

I have done myself a bunch of graphs that I think paints a fairer picture of Autodesk’s position, but there’s a reasonable chance I’m wrong about that too so I won’t be publishing them. Instead, In a day or two, I will remove the content of the offending posts (but leave the shell of the posts there to preserve the comments). I do this not to hide my embarrassment, but to limit the degree of undeserved damage to Autodesk. Feel free to copy/paste, take screenshots, etc. of the posts until then. Of course, it’s not really possible to delete things from the Internet, so if you ever want to relive the joy of seeing me get things spectacularly wrong, feel free to use the Internet Archive to do so.

What this doesn’t mean

This doesn’t mean Autodesk is off the hook with the rental thing. It may not yet be a financial disaster of the magnitude I argued, but the jury is still very much out on whether it will eventually succeed. Even if it does (and I still have very strong doubts – doing the opposite of what your customers want is rarely a winning long-term strategy), it’s still a grotesquely anti-customer move which deserves to be vigorously opposed. I will  continue to oppose it. I will continue to point out any faulty arguments that are used to support it. However, I will be much more careful to avoid using faulty arguments of my own.

* Autodesk could have also pointed out my error, but didn’t. Before I started commenting on rental I emailed Autodesk PR specifically encouraged them to point out any factual errors and/or seek a right of reply, but I want to make clear that isn’t what happened here. I have not backed off due to pressure or threats from Autodesk. Indeed, I have had no contact from Autodesk whatsoever in relation to my blog since it re-started. I continue to encourage such contact, but of course Autodesk is under no obligation to take up my offer.

Battle of the Bullshit part 2 – Autodesk’s sophistry

In my last post, I gave Bentley a well-deserved slap for, er, saying things that perhaps weren’t entirely factual. Now it’s Autodesk’s turn.

What’s this about? Carl White, Senior Director of Business Models at Autodesk, wrote a blog post Not so fast Bentley: Separating fact from fiction responding to statements made by Bentley in its press release Bentley Announces Autodesk License Upgrade Program. Some of Carl’s observations on Bentley’s claims were perfectly valid, but unfortunately he went beyond that and wrote a few more things – “facts” – where he’s on shakier ground. Let’s examine Carl’s interpretation of reality, shall we?

Fact #1 – No Autodesk customer ever  loses the right to use the perpetual software license you’ve purchased, it is “evergreen”.

This is generally true. There are exceptions (read the EULA), but let’s not split hairs. In the vast majority of cases, we don’t lose the right  to use the software. We can, however, lose the ability  to use the software. That loss is practically inevitable long-term because of the progress of technology. I have several old AutoCAD releases I can’t run for environmental reasons, not licensing ones. This means that if we want to use our licenses long-term, we rely on Autodesk’s ongoing cooperation. That’s where customers have legitimate concerns, because there are no guarantees that Autodesk will continue to provide that cooperation. If it does, there are no guarantees that cooperation will remain free or even affordable.

And if you’re on a software maintenance plan, you can continue to receive all of the benefits of software updates and technical support for as long as you’d like.

This has been officially promised, and let’s give Autodesk the benefit of the doubt and assume that this promise will be fulfilled to the letter. There’s still an elephant in the room. What will the benefits of updates and support cost us? Based on what Autodesk has done in recent years, it is a pretty safe bet that the cost of maintenance (formerly called Subscription) is going to rise, and rise sharply. Give it a few years and I expect maintenance customers will be paying the same as rental customers. I expect other strong-arm methods will be used to “encourage” people onto rental. When this happens, our perpetual licenses will be near worthless and Bentley’s claim about a “…write-off of the future value of their investment…” will become uncomfortably close to the truth.

We’ve shared key dates well ahead of time to give customers time needed to adjust, but that does not mean we’re taking away options.

The latter part of this statement goes beyond disingenuous; it’s arrant nonsense. Of course Autodesk is taking away options. Autodesk has been taking away options for years, and this has only accelerated. As of right now, I can no longer buy an Autodesk software perpetual license. I no longer have that option, which I had before. How is that not taking away options?

Fact #2 – Our customers have a choice. When you subscribe to Autodesk software, you have flexible terms (monthly, quarterly, annually), and multiple access points (single user, multi-user and shared). Now Autodesk customers can get the software they need for a year or a month, in ways that are more convenient and better for their business.

Well, I guess the first sentence is kind of true in a sense. Long-term customers (that’s most of us) do have the choice between paying merely a lot  more per annum for an Autodesk license via annual or multi-year rental, or paying vastly  more by doing it monthly. Suggesting this is better for our business is, of course, laughable.

Customers can buy and use it for as long as they want and can match their subscription type with the demands of their workforce. When the workforce expands, they can ramp up, or in quieter periods, they can scale it back. In short, subscribing gives you flexibility and predictability.

This is true; rental is  the best option for some customers under some circumstances. It is good that Autodesk has made that option available for the small minority of customers in that situation. However, it is the opposite of flexibility to make it the only  option.

When it comes to value, lower upfront costs make our software more accessible and allow you to try more tools without the risk of a large upfront expenditure. Plus, you only pay when you need it. This is a big deal. Some of our customers prefer this cost is considered an operating expense, allowing you to bill the cost of the software back to the client or project. And if you subscribe for a longer, multi-year term, you lock-in your rate. Combine that with flexibility in the length of contracts and you may find that you’re actually paying less.

Nice attempt at spin here, but ultimately it’s nonsense. Except for the minority of customers who need that level of flexibility, rental is not about paying less. If it was, Autodesk wouldn’t be doing this. Pushing Autodesk customers on to rental is all about trying to extract more  funds from us for the same thing, not less. Suggesting otherwise is disingenuous.

If rental really was  better value, Autodesk would give its customers the choice between perpetual and rental and let the market decide. But wait! Autodesk did exactly that a few years ago, and the market decided; the rental experiment failed miserably. Autodesk knows  it has to make rental compulsory because otherwise most customers wouldn’t go for it. Yet in a painful piece of patent paralogy, it paints this compulsion as a selfless act of customer service.

Fact #3 – Software as a service is essential for technological evolution. It allows for continual and consistent innovation and support. The software will get better, faster and more seamless in the way you use it. The experience is customized to you or your organization, and provides a simplified way to access and deploy software, manage your users and collaborate on projects. With this new way of delivering software, everyone will always have the latest, most up-to-date Autodesk tools available.

Even ignoring the conflation of software as a service (SaaS) and rental, the first sentence is breathtaking in its audacity. It goes beyond spin, beyond disingenuity, into the realms of the surreal. No, SaaS is not essential for technological evolution. The whole history of computing screams that loud and clear. Autodesk wouldn’t exist if the first sentence were true. It isn’t remotely close to true. To be generous, it’s a terminological inexactitude.

Reading beyond the first sentence, there’s a lot of wonderfully utopian wishful thinking that nobody familiar with Autodesk would believe for a second. It’s shown up for the other-worldly spin that it is by Autodesk’s years-long ongoing decline in maintenance value-for-money and its woeful attempts at trying to make continual updates work (which you probably don’t want anyway).

It’s not just Autodesk saying this; the entire software industry is moving in this direction. Frankly, design and engineering software has been a bit slow to make this change. But the benefits for end-users are clear, and it’s just a matter of time before all vendors have similar ways of buying.

While it’s true that various software companies are moving at least partly towards SaaS and rental of conventional software (some more successfully than others), it’s not at all a uniform industry-wide position. It’s disingenuous to imply that going all-rental is already almost universal and Autodesk is just catching up. As for the “benefits for end users” being clear, I guess all those customers who like paying lots more per year for their software will agree.

Yes, it’s likely that many vendors, maybe even most of them, will have similar ways of buying in the next few years. No, it won’t be all of them. No, not all vendors will make rental compulsory for new licenses as Autodesk has done. Some of Autodesk’s competitors (e.g. Bentley, Bricsys) will continue to provide their customers with the ability to purchase perpetual licenses. The law of give-the-customer-what-they-want-or-die tells me that those competitors are much more likely to thrive than Autodesk.

What does this mean? It means that millions of you are already seeing the benefits of shifting to subscription and are making that choice voluntarily.

Voluntarily? Really? I can’t imagine anyone typing that statement in that context without either wincing (if they have any self-respect) or laughing (if they don’t). Strewth!

So who won the Battle of the Bullshit? Nobody. First, Bentley lost. Then Autodesk put in a supreme effort, summoned up a steaming stack of sophistry, and lost more.

Raise your game, people; we’re not all stupid out here. If you can’t support your argument with the truth, then your argument isn’t a good one and you need to rethink it.

Battle of the Bullshit part 1 – Bentley’s terminological inexactitudes

I note with interest the blog post Not so fast Bentley: Separating fact from fiction by Carl White, Senior Director of Business Models at Autodesk. In this, he responds to statements made by Bentley in its press release Bentley Announces Autodesk License Upgrade Program, stating:

Earlier this week, Bentley announced an “upgrade program” for Autodesk customers. We found the offer to be disingenuous and mischaracterizes what Autodesk offers our customers.

OK, let’s have a look at what Carl is complaining about. Here’s one Bentley statement that could be considered questionable:

For consideration by owners of Autodesk perpetual licenses facing Autodesk’s imminent deadline for the write-off of the future value of their investment, Bentley Systems is offering recovery of the value otherwise subject to forfeit.

Carl has a point here. The “imminent deadline for the write-off of the future value” line is presented as fact, but at this stage it’s not true. While perpetual license owners may legitimately fear for the long-term value of their investments, there is nothing subject to an imminent deadline other than the end of the ability to purchase further perpetual licenses. Likewise, the “subject to forfeit” thing is a scaremongering phrase that deserves Carl’s “disingenuous” label. Autodesk isn’t subjecting anything to forfeit right now. Anything else dubious in Bentley’s statement?

Bentley Systems considers purchases of perpetual licenses to be long-term investments by our users, so we continually innovate to increase their value. We are glad to now extend this ‘future-proofing’ to Autodesk license owners who otherwise will lose value in their applications.

That’s all pretty reasonable but the “…will lose value in their applications” part is questionable. We might suspect that will happen, but we don’t know  it yet. Perhaps “…may  lose value in their applications” would be more reasonable. Bentley also quotes a customer as saying:

Autodesk continually sets deadlines forcing us to give up our perpetual license for an annual subscription.

Now while it’s accurate to say that Autodesk continually sets deadlines and has certainly been very heavy-handed in its years-in-the-making push to rental (currently called subscription in Autodeskspeak), it has not yet forced customers to give up perpetual licenses. Those of us with perpetual licenses have not  been forced to give them up. We can continue to use them. Bentley shouldn’t use inaccurate statements like this in its marketing, even when quoting others.

In summary, Carl is right. Bentley has  been disingenuous and deserves a slap for it.

If only Carl had just stuck to the sort of analysis I made above, I could have ended my own analysis right there. Unfortunately, he didn’t. He couldn’t resist the urge to add his own “facts”. My next post will put these under the same kind of scrutiny.

AutoCAD 2017 Service Pack 1 is out but you probably don’t want to install it

As reported by Jimmy Bergmark, AutoCAD 2017 SP1 will break add-ins that use Autodesk’s built-in autoloader mechanism. It looks like it’s a problem caused by third party applications, but it’s not. It’s entirely Autodesk’s fault. The only fix at this stage is to uninstall SP1.

It’s astonishing that Autodesk would release a service pack like this, introducing a nasty bug that will break customers’ existing functionality. This reminds me of the comedy of errors that was AutoCAD Release 13 with its multitude of updates, many of which introduced new bugs as well as fixing others. AutoCAD 2017c4a, anyone?

If you needed any more evidence that automated continuous updates from Autodesk are A Bad Idea, here it is. What a crock.

Autodesk desktop app. Worst. Name. Ever. Is the product better than the name?

Autodesk wants your software to be automatically updated so you’re always running the latest version. Let’s pretend for a moment that this is a good idea and have a look at how Autodesk now attempts to do this. For the previous couple of releases (2015/2016), this has been done using Autodesk Application Manager. For 2017, this has been replaced by Autodesk desktop app. Even if you haven’t installed any 2017 products, you may have already seen this kind of thing pop up. Repeatedly.

ApplicationManagerReplaced

Note how there’s no obvious “stop nagging me and leave me alone” option. Autodesk Application Manager’s settings page does include an Alerts tab which allows you to turn off all desktop alerts, but the above message indicates Application Manager has suffered an “end of life” experience so there’s not much point having it on your system.

Before I get onto the new product and how it works, I want to discuss its name. It’s woeful. I have never been less whelmed by any product name than Autodesk desktop app. It’s not even Autodesk Desktop App, it’s Autodesk desktop app (sans initial capitals). It’s dull, generic, uninspired,  and it means nothing. AutoCAD is an Autodesk desktop app. Inventor is an Autodesk desktop app. Graphic Impact used to be an Autodesk desktop app. This is the equivalent of Ford naming their next new car “Ford road vehicle”. It says to me, “We couldn’t come up with a name so we just gave up.”

Having got that off my chest, what about the product itself? What does it do? According to Autodesk, this:

Autodesk desktop app is a content delivery solution. The desktop component installs with Microsoft Windows®-based Autodesk 2017 products and suites. It replaces the previous in-product update components and the Autodesk Application Manager.
 
Autodesk desktop app keeps Autodesk Subscription customers informed of product updates, new releases, new features, and special subscriber-only learning and training materials, as they become available. Autodesk desktop app also delivers and applies critical security patches for all 2015 Autodesk products onwards.

Does it deliver? No. Not for me, anyway. In my secure proxy server environment, it fails to connect, despite Autodesk Application Manager working fine in the same environment. There is a long and unfortunate history of various aspects of Autodesk software switching between working and not working in a secure proxy environment, and this product just added to that history. Things that worked in one release stop working in a new release. Some of them might start magically working again in a later release. I can only assume it’s developers tinkering with stuff and not realising the consequences. Whatever the reason, it inconveniences customers, makes Autodesk look bad at the Internet (again), and throws a spanner in the works as far as Autodesk’s online-driven future plans are concerned.

Besides it having the dumbest name ever and (for some) not doing the one thing it’s designed to do, what else is wrong with it? Here’s a list for starters:

  1. After an AutoCAD install, unless you specifically choose not to install it, it will be installed and will automatically run, getting right in your face when you probably just want to draw things.
  2. It automatically sets itself up to run all the time, using your system resources. I intend to go into more detail about Autodesk web stuff abusing your resources in future.
  3. On systems where it doesn’t work, the user experience is, er, sub-optimal. I won’t bore you with the details, but you are sent on a wild goose chase.
  4. On systems where it works, it calls home and communicates with Autodesk in undocumented ways (“Collects usage data to support users better.”).
  5. On systems where it works, it may repeatedly tell you about updates you have already applied.
  6. On systems where it works, it may not keep you up to date in the advertised manner; some hotfixes are not supported.
  7. It has a Sign In button, which encourages you to get an Autodesk account if you don’t already have one. It should not be necessary to sign in to anything to keep your products up to date. If you’re managing a corporate environment, you might not want your users all signing up for individual accounts without your knowledge.
  8. The whole idea of giving Autodesk unfettered access to update your apps when it feels like it is pretty horrific. That would involve a naive level of trust in Autodesk that it has proven repeatedly it does not deserve. The very application that wants to auto-update so it can carry on doing auto-updating other things is itself a classic example of why you really don’t want to allow that sort of thing to happen.

Let’s say you don’t care about any of that and still desperately want this thing working. What to do? At the bottom of the About Autodesk desktop app knowledgebase article, there is a link: Autodesk desktop app is not able to connect to the Internet. When I clicked it, I got this:

ThisIsEmbarrassingWell….this really is embarrassing. To be fair, the link appears to be working now, so you probably won’t see the above. However, what you do see isn’t any better. The advice provided there is terrible. It’s asking for totally unreasonable holes to be punched through your firewall, allowing whitelist access to a myriad of marketing and other undesirable sites (e.g. Akamai, which has a history of sub-optimal online ethics). I don’t think so.

The best thing I can say about Autodesk desktop app is that you can choose to not install it with your 2017 product. That seems like an excellent idea to me, because this product is a crock. Uninstall it if it’s been installed. Uninstall Autodesk Application Manager if that’s installed, because it’s just going to stop working and nag you to death in the meantime. Then store away this little episode at the back of your mind for future reference, because I’m sure Autodesk hasn’t finished pushing us down this path yet. Push back. Hard.

AutoCAD 2012 – Massive download bloat

Note: this post is not an April fool’s joke. It may be ridiculous and hard to believe, but unfortunately it’s all true.

After I managed to overcome Autodesk’s obstructive download manger and download AutoCAD 2012, it became available on the Subscription site (when that site wasn’t running unusably slowly). Or it became kind-of available. Here’s what is actually available:

  • AutoCAD 2012 Multilingual 32 bit
    Download File Size: 2,080,558,319 bytes (1,984.2 MB)
  • AutoCAD 2012 English Korean Traditional-Chinese Simplified-Chinese Win 64bit
    Download File Size: 2,240,915,999 bytes (2,137.1 MB)

These file sizes are roughly double those of the AutoCAD 2012 English files I’ve already downloaded from the trial page and installed. The 32-bit English file is 1,144,011,680 bytes, or 55% of the size of what the Subscription site is trying to offer me.

Why? Because the Subscription downloads contain three bonus Asian language packs. It has apparently escaped Autodesk’s notice that Australia is an English-speaking country, and that the ability to install a Korean version of AutoCAD 2012 isn’t going to be spectacularly useful here. Duh!

There was a a distribution fiasco last year when Autodesk couldn’t make up its mind which AutoCAD 2011 language variant Australian users were supposed to use. This resulted in weeks of delays, uncertainty and disrupted shipments. This year, there’s less uncertainty. Somebody has made a firm decision about what we’re getting, right from the start. What a shame it’s the wrong one, and it makes Autodesk look utterly clueless.

Just in case you’re wondering, the AutoCAD 2012 English from the trial page installs and works fine, correctly detecting that I’m in Australia and presenting the correct legal information. The installation also registers and authorises correctly using the serial number provided on the Subscription site. No problems there, then.

What, then, is the reason for the massive download bloat? Is it really just stupidity, or is there some legitimate reason for it? I’m informed that installing the English version of 2012 in Australia may cause some problems with Migration when upgrading to 2013. I am unconcerned about this for two reasons. First, I’m sure migrating from 2012 English to 2013 English will work just as well here as it does elsewhere in the world. it’s not as if the Migration utility has to invert the bits or anything for Down Under users. Second, I have avoided Migration anyway since AutoCAD 2006, when “improvements” rendered it effectively unusable to me.

More Autodesk deception over LT productivity study

Following on from the AutoCAD 2011 productivity study I critiqued earlier, there is now an LT version. Do the same credibility problems apply to this study too? Yes, and then some.

In addition to the drawings and operations being deliberately hand-picked to demonstrate new features, no direct comparison is performed at all between the two releases on the same platforms. Every single quoted “productivity improvement” figure includes, free of charge, three years of hardware and operating system progress and a more upmarket graphics card.

If you read business “news” sources that just reprint press releases, such as this Yahoo! Finance one (thanks, Carol Bartz), you won’t see this mentioned. Instead, you will see deceptive statements like these:

David S. Cohn, an independent consultant

Er, no, in this context he’s not independent, he’s an Autodesk consultant. Not that there’s anything wrong with that.

overall productivity gains of 44 percent for users moving from AutoCAD LT 2008 or earlier versions to AutoCAD LT 2011

…as long as you only ever perform certain carefully selected operations and upgrade your hardware and operating system. Like the other study, the 44% figure is totally meaningless and quoting it without qualification is downright deceptive.

Most users will be able to get more work done faster by upgrading to AutoCAD LT 2011

This statement is totally unsupported. There is no analysis of what “most users” do with the software, and no attempt to quantify the portion of time such users spend on these hand-picked operations. Neither is there any analysis performed on more common operations to see if the new releases introduced any detriment to productivity in those areas.

Improvements to the graphical user interface deliver a 43 percent productivity increase.

If that’s true, why do so many users of 2009 to 2011 immediately turn off the new user interface? Are they all stupid Luddites who have a burning desire to work much less efficiently? This study, like its non-LT counterpart, contains many unqualified statements about the Ribbon improving productivity and providing other benefits. I’d really like to see a proper independent study done into that.

To sum up, Autodesk is quite prepared to say misleading stuff about its products that will be regurgitated unquestioningly by those who don’t know any better, in the hope that it will be believed by those who do, and not exposed by those who care. But it’s not prepared to answer straightforward legitimate questions about its business, offering a pile of spin instead. This, supposedly because “management in publicly trade companies are forbidden by US laws and accounting regulations to discuss some topics”.

I think I’ll borrow a phrase from Deelip here, as it seems appropriate.

Bottom line. This is bullshit.

It just so happens that right now I’m in a no-bullshit mood. I’ve been exposed to more than enough of it lately. Unfortunate timing, Autodesk.

I know this sort of marketing device is nothing new, and maybe that’s the point. This kind of thing is so 20th century. In the good old days, negative commentary about stuff like this would be seen by few, and largely confined to company-controlled environments and one-way media such as printed magazines. Things aren’t like that any more. This sort of nonsense is being increasingly noticed, criticised and derided in blogs and social media. I have hope that the point will soon come when companies’ PR consultants work out that the negatives of spewing bullshit outweigh the positives. When that point is reached, the bullshit will stop. And won’t that be great?

Magic vanishing images

In a thread in the Feedback & Questions about the Discussion Groups section of the Autodesk discussion groups, somebody called ACADuser contributed what I thought was a highly amusing bar graph as a test image. Inspired by this, I contributed a couple of test images of my own.

A few hours later, the whole thread magically disappeared! It seems a shame that I went to the effort of making those images, and all for nothing. The handful of people who would have seen them on the discussion groups have now missed out on the experience. So I’ve decided to make up for that by posting them here, where thousands of people can look at them instead.

Here’s the first one (not that amusing):

Discussion group upgrade poll results

Here’s the second one. Given the circumstances, it seems somewhat prescient:

Clue train pie graph

If ACADuser wants to get in touch, I’ll be quite happy to post his image too.

Autodesk discussion group changes – user reaction

I will be airing my own views on the Autodesk discussion group changes in a future post. In the meantime, I have collected some reactions from other users. For the record, there has been only a little censorship in this area. Here are some of the comments that made it through unhindered:

  • I’ve given it a fair shake and it’s just as bad as I imagined
  • Goodbye, people. It was nice while it lasted
  • it sucks
  • it doesn’t look like you have any intention to meet the expectations of these people
  • not [as] much traffic as there was before the change.  I hope things improve
  • I’m sure you’ve noticed the sourness many folks are having with this interface
  • What a f’in f-up
  • This is so aggravating that I am resorting to posting questions that may have already been answered vs. trying to find them via the search tool
  • Very annoying
  • We use NNTP because it’s easy and fast, and better
  • very slow, compared to “other” html forums
  • This was hyped as a “state-of-art web experience”. It is clearly not
  • Extremely slow compared to the previous web forum
  • we are screwed with this interface
  • This is like having your high performance vehicle (NNTP) stolen and having to take the bus to get to your destination
  • 4 days later, still sucks
  • Still very slow, cumbersome, difficult to track and navigate, unintuitive
  • It took me literally 30 seconds to get that smiley to insert
  • I really was expecting something better
  • I see too many people who may not be around anymore. In most cases their expertise far outweighs any improvements to the forums
  • Welcome to the new and improved Autodesk forum brought to you by high school students near you
  • You keep using that word [“upgrade”]. I do not think it means what you think it means
  • Better? Wanna bet? It’s cumbersome, at best
  • Another annoying thing here is that I can’t seem to find a way to show threading
  • Very, VERRRY slow, compared to forums using PHP and the like
  • Why even have an edit feature at all, when it’s virtually useless?
  • the “experts” are being alienated and having a harder time contributing to aid the beginners
  • I don’t like reading this forum in a browser. NNTP was and is much better
  • This sucks
  • about 40 unanswerable questions that have popped up in the first 30 (wasted) minutes of trying to “give it a chance”
  • If we aren’t going to get our newsgroups back at least attempt to make this forum professional
  • Sorry folks but I just don’t have the time to log in and browse thru all the different pages required now
  • Map 3d is “losing” its best contributor because of a dumb forum update
  • Autodesk prove again if something works  they will find something to make wrong
  • I stopped posting here after the change for the same reasons.  Just logged in for this
  • since the demise of the NNTP feed I rarely visit several of the forums I used to watch
  • It is just too time consuming now. This is really discouraging
  • I cannot be nearly as productive as I could with a newsreader…it takes no less than 4-5x longer
  • getting rid of the NNTP server was one of the worst things Autodesk has done in years
  • Goodbye
  • your update and support policy really force me into alternatives to Autodesk

That’s going down well, then. To be fair, there have been a few people who are relatively supportive of at least some of the changes. As usual with any unpopular change, there are a couple of asinine comments attacking the critics as just a bunch of old whiners who are resistant to all change. But the selection of comments above reflects the overwhelming negative sentiment, and that’s from those people who bothered to stick around long enough to make their views known.

I would have thought Autodesk would have learned its lesson after the well-deserved thrashing it got the last time round, but apparently not.

AutoCAD 2011 online Help changes – a curate’s egg

As announced by Shaan Hurley, Autodesk has made some changes to the AutoCAD 2011 online Help system. Please check it out and see what you think. After a short time with it, here are my experiences using IE6 (yes, I know). As this is a dynamic system and dependent on browser characteristics, Internet connectivity and any changes Autodesk may make between me writing this and you reading it, your mileage will vary.

There are some cosmetic changes,  including a fixup of the Autodesk logo in IE6 that was done a couple of weeks ago. Sadly, my pink Comic Sans logo has not been adopted.  As I can’t do a direct side-by-side comparison with the pre-change setup under identical conditions, I can’t make a definitive statement about performance. I can say that it does appear to have improved somewhat. It now takes about 3.5 seconds from hitting F1 to seeing a complete landing screen. Once cached, I’m seeing it in come up in just under 2 seconds.

The main change from a usability point of view is that the Search facility now defaults to searching All Books rather than whatever document you happen to have highlighted over on the left sidebar. That’s welcome. Also, the searches generally appear to give better results. For example, a simple search for LINE in the original 2011 online system gave a list of 199 results, of which the actual LINE command was 26th! Now, a search for LINE puts the LINE command third in the list; much better. The results come up faster than before (2.5 seconds in this example), but I have seen widely varying search times reported so I would be interested to hear about your experiences.

The way the search results are presented is now significantly different. Instead of a single line for each result, 4 lines are now used. There is a descriptive hyperlink line, a line containing sample of text from the page the link points to, a spelled-out link line and a blank line. The second line appears to be randomly chosen. In our LINE example, the text starts with “If the most recently drawn object is an arc…”, which is a fairly long way down the LINE command page itself. The third line’s only function appears to be to waste space. I can right-click on the main link if I want that information. I can’t even copy and paste that third line; attempting to click and drag in the search results pane selects the whole lot.

Because of the newly verbose display format, it obviously doesn’t make sense to display 199 results, as there would be too much scrolling. What is now displayed is 8 results, with links to another 7 pages (I’m not sure what happened to the 199 – 64 = 135 other results). If your desired result is in the first 8, that’s fine. If it’s not, then you have a harder job now to find what you’re after. You can’t use your browser’s find feature to look for a specific word among the full set of results. You will have to click on each page in turn and wait for it to appear before scanning the results. Fortunately, each page comes up fairly quickly (about 1 second), but I would much prefer to have the option of seeing more results displayed on each page. I suspect things have been arranged this way to improve performance (fair enough) and make it work better on mobile devices. While that’s all nice and cool and trendy and geeky and everything, I don’t intend to ever use AutoCAD on an iPhone. I would much prefer it if Autodesk prioritised its user interface design based on what the vast majority of its users are going to be using when they need the documentation.

Choosing a different page within the search results and then using the Back button takes me back to the main landing page rather than my previous results page. Using the Forward button to try to get back where I was, just puts me back on page 1 again. This is obviously not good.

That’s enough of the changes in isolation, how does today’s system compare with what went before? I did a quick test to see what was involved in finding out about a given command. I chose the WBLOCK command. Other commands and other users may give better or worse results.

AutoCAD 2010 CHM Help
Method: F1, type W, double-click on WBLOCK
Time taken: 2.6 seconds

AutoCAD 2011 offline Help (as shipped)
Method: F1, click in search box, type W, Enter, click on W commands, click on WBLOCK, click on Write Block Dialog Box
Time taken: 12.1 seconds

AutoCAD 2010 online Help (as at 12 May 2010)
Method: F1, click in search box, type W, click outside search box, click on search arrow, click on W commands, click on WBLOCK, click on Write Block Dialog Box
Time taken: 16.0 seconds

Given those results, it would be pretty hard to argue that the new system is more efficient for users. Again, this is just a sample command and method, and if you can find a different one where the new stuff works better than the old, I’m all ears. The method I ended up using for testing the 2011 online search actually required a fair bit of trial and error. Here are some things I tried first:

  1. F1, click in search box, type W, autocomplete gives me ‘wireless’ (left over from some other search I used on an unrelated site), Enter, nothing happens
  2. F1, click in search box, type W, click outside search box so I just have W, Enter, nothing happens
  3. F1, click on search arrow, click in search box, type W, it gets added to the end of the word Search, giving me ‘SearchW’!

A few minutes after my tests, I tried again to see if there was a better way. What I found was that the W Commands link I needed was completely absent from the search results!

Look, no W Commands!

W System variables, check. W Methods, check. W commands? Nope. Not on this page. Not on any of the other 7. Where did it go? Will it return one day? Who knows?

One of the risks of online-based software is that it can be a moving target. Stuff that you used in the past may not be there the next time you need it. It’s easy to see users getting confused and frustrated by this kind of stuff. After all, it’s supposed to be Help, not Hinder.

In summary, some of the changes are welcome, but the system is still a long way short of being anywhere near as efficient or friendly as the one it replaced. The performance is better than it was, but still slow. The interface contains some clangers that tell me that user feedback has been absent, inadequate and/or ignored.

I suggest this system be withdrawn, and soon. AutoCAD 2011 Update 1 should contain a complete and properly integrated CHM-based Help system, and Autodesk should go back to the drawing board with the whole browser-based Help idea.

If, after due consideration and extensive user consultation, Autodesk still thinks that online Help is a good idea, it should spend the time required to make it work properly, introduce it only when Beta testers are satisfied that it is at least as good as what it is replacing, and then introduce it alongside the CHM system. The two systems should be run in parallel for as many releases as it takes to convince the vast bulk of users that online is best, at which point the losing system can be discarded.

Right now, it’s abundantly clear which system should be ditched, and it’s not the CHM one.