• Do Autodesk customers need to be on Subscription to receive bug fixes? No, they do not.
• Do Autodesk customers need to have purchased the software within the last 90 days to receive bug fixes? No, they do not.
• Do Autodesk customers need to have reported certain specific bugs to receive bug fixes? No, they do not.
• Do Autodesk customers even need to be running the current release to receive bug fixes? No, they do not.

AutoCAD 2010 Update 2 (that’s Service Pack 2 in the old language) has just been released for the users of last year’s software. This includes the Update 1 changes. The usual caveats apply, including reading the Readme first. As usual, Autodesk’s oddball numbering system means that after installation, Update 1 shows up as Version 2 and Update 2 shows up as Version 3 in the About screen.

This Update applies to straight AutoCAD (and LT), not the vertical variants. I have no news about non-English versions. Patrick Emin informs me these updates are language-independent.

Please add your comments!

Autodesk FM Desktop suffered this fate, and if you go looking for information about the product on the Autodesk site you’ll find only a few dregs left over from the days when this was a viable product. At least in this case Autodesk has belatedly arranged a path out of the mire for its customers. FM:Systems will be taking over Autodesk’s FM customers, and your Autodesk FM Desktop licenses can be converted to FM:Interact Workplace Management Suite licenses. There is no data transfer mechanism yet (other than a DWF import), but something is supposed to be made available in the next few months.

(Edit: Marty Chobot from FM:Systems informs me that they will help customers who need to transfer data immediately).

For more information, see www.fmsystems.com/fmdesktop, especially the FAQ.Edit:

Now I happen to know that Autodesk goes to some lengths to find out what its customers are thinking (more on that later), but still this feeling of being ignored persists among its customers. Why is this so? Why do so many of you hold this view? I have my own thoughts about this, but right now I’m more interested in yours.

I’d like to see some examples that make you think that Autodesk doesn’t care about your viewpoints, wishes and desires. If you can suggest ways in which Autodesk could do things better, let’s hear them. On the other hand, if you believe that Autodesk is listening, please provide examples that show that to be the case.

I will be asking some Autodesk people the same question very soon, in addition to any questions you pose in my Ask Autodesk a question post (more questions, please!). I will report back on the answers, so it will be interesting to compare your comments with what the Autodesk people have to say.

Here are the changes I see:

• The old messages and Plain Text new messages that had their paragraphs stripped out have had them returned. This is the end of the good news, as far as I can tell. If you only ever like reading positive things, particularly about Autodesk, I suggest you stop reading now.
• Rich Text messages have had superfluous paragraphs introduced, and other formatting issues. The more you edit a message, the worse it gets. Try switching back and forth between Rich Text and Plain Text a few times, it’s a crock.
• My email address has been newly exposed to the spambots as my user ID, and I’m not alone. WTF? Totally unforgivable! Heads should have rolled the first time this happened. Doing it again is way, way beyond endurance.
• The Edit Account page refuses to allow me to make changes, falsely claiming that several fields contain incorrect data. This means I can’t fix the email exposure problem myself, and have to wait for Autodesk to fix it. Never mind, I’ve done a lot of waiting for Autodesk over the past few months, so I’m kind of used to it.

Other than that, the discussion group system looks like the same old pile of garbage that it has been for weeks. For example, Search is still broken. Paste is still arcane. Using the indent or bullet items in Rich Text results in text above the selection being indented too. The question evaluation system (of dubious value in any case) causes threads to start with such nonsense as “This question is not answered. Helpful answers available: 2. Correct answers available: 1.” Er, OK then…

Even the Help page hasn’t been updated to reflect any alleged new fixes or to correct any of the errors I pointed out weeks ago.

All in all, I’d say it’s not exactly a major triumph as a maintenance effort.

Autodesk, please give up trying to maintain this steaming pile of pus. It’s not going to work. If the people at Jive Software can’t help you get their forum software working (and there has been plenty of time in which that could have happened), then bite the bullet and throw it away. Really.

Your continued patience is appreciated as we work to resolve the discussion group issues you have been reporting. We understand the impact these issues have on your productivity, and want to assure you we are continuing to troubleshoot and resolve. We’ve posted an update under “Help” to provide awareness and status of the issues we are working on. We’ll regularly update this as improvements are made.

Never mind the glacial nature of the response, it’s good to see that an acknowledgment has finally been made of the problems. However, picking on the Help link reveals that there’s a long way to go yet before all the problems are even fully understood by the team responsible, let alone fixed. Only three “Known Issues” are listed, and four issues are allegedly resolved. At least one of those, shown as resolved on 7 October, is still very much broken right now. At least one of the FAQ items, “Why can’t I stay signed in?”, gives false information.

Discussion group team, you will find a lot more than seven issues listed on this blog alone. To see them, just click on the Newsgroups link in the Tags section on the right. Alternatively, you could use the Search box at the top and enter something like “discussion groups”. A search that actually finds everything? There’s a novel idea.

Now, my family and I had pretty much arranged our whole weekend based on having no power on Saturday morning, postponing various things such as housework and washing to fit in with the time we had been given. When he hadn’t turned up by 11:00, it was time to think about preparing the kids’ lunches before he turned up. He hadn’t turned up by 2:00 when I had to take the kids out somewhere, and he had given up answering his phone, so I just went out. He hadn’t arrived when I returned, and still hasn’t arrived, so I think it’s fairly safe to say he’s never going to turn up.

You might think that this piece of total non-service represents some kind of low point, but it’s actually only the third worst customer service experience I’ve had this year. I have had two much worse experiences than that in 2008; one from an Australian company and one from an American company. I will relate those experiences in later posts.

Anyway, my customer service experiences, along with some other recent happenings, have led me to contemplate such matters, and come up with some (ahem) profound thoughts on the subject. The main point is that getting things right is only the first line of defence against poor customer service. What happens when things go wrong (and they will) is the real test of a company’s service attitude / culture.

What should happen when things go wrong? Three Fs. F— up? Fess up. Fix up. What do these mean?

• F— up? Find out if a mistake has been made. If a customer is saying that something has been screwed up, assume they are right until proven otherwise. Even if complaining customers are wrong 90% of the time, that’s no excuse for treating the other 10% as if they don’t know what they are talking about, or that they are wrong or unreasonable.
• Fess up. If a problem is evident, admit it. I can think of no single instance in corporate history where denying the existence of a genuine problem has made a company look better than admitting it. Ignoring a problem, making excuses for it, obfuscating, pretending it’s not there, or even claiming that it’s the customer’s fault, always makes the guilty company look worse. Always. No exceptions. A company representative that acts like that is doing no kind of useful service to the company, no matter how loyal they may think they are being. The same thing applies to politicians, but don’t get me started on that.
• Fix up. Having established that the problem is real, correct it. Do whatever needs to be done to make the customer satisfied.

I could add a fourth F, “Fast”, because correcting problems that affect customers should be a priority, but let’s keep it to three for now. Easy, right? I expect I will now make a fortune writing a book, selling my ideas to clueless companies that should know better, and/or doing highly paid speaking tours. I think I will call it, “The Johnson Method of Customer Service: It’s F—ing Obvious.”

I was a little worried that when one part of it eventually failed, I would be stuck with a partially functional device, such as a scanner/fax that wouldn’t print, or a printer that wouldn’t scan, and be left with the dilemma of replacing all of it or part of it. But I had good experiences with long-lived printers in the past (albeit Hewlett-Packard ones), so I figured that if I had to throw it away in five years’ time I could live with that.

In practical use, most of the device’s features turned out to be as advertised, and while it was working I was generally happy with it. But I won’t be buying another one, and it’s unlikely that I will ever buy another Canon printer of any description. Why not?

• Performance. This simply isn’t up to scratch. While it may theoretically print a 500-page document at 30 pages per minute, printing a single page is a different matter. Although it can look spectacularly quick in action, it takes one full minute from turn-on to get itself ready to do anything at all, then about 10 seconds to print a simple monochrome page in draft. There are also long delays when the device is switching from one kind of use to another. The lengthy period of whirs and clunks indicates that it’s doing something very important internally, but I have no idea what. I don’t care. For my typical use, it’s just too slow.
• Economy. The ink savings promised by the 5-tank system are illusory. This thing eats ink at a rapid rate, so I’m finding that the costs of running this printer are significantly greater than my previous Hewlett-Packard. Having to maintain at least one spare (preferably more, because they don’t last long) of each tank is inconvenient and means there is always an expensive set of tanks lying around waiting to be thrown or given away when the device finally dies. Which, given my experience to date, could be any day now.
• Reliability. It doesn’t have this. It had to be returned for warranty repairs in its first year, as it complained about its ink tanks. This resulted in the print head being replaced. Out of warranty, it started doing the same thing again. This was sometimes fixable by various means, such as removing and replacing the tanks, switching the device on and off, removing and replacing the print head, cleaning the contacts, prematurely replacing unfinished ink tanks with new ones, and so on. This would sometimes fix the problem on the first or second attempt, but this level of cooperation didn’t last for ever and the condition gradually worsened until the device was officially dead. I took it in for repair but apparently a new print head (which costs 30% as much as the printer) was not required this time. It has been fixed, for now, by replacing one of the half-full tanks with a new one. Apparently, genuine Canon tanks, which are the only thing it has ever had in it, are prone to bad batches, and I’ve been unlucky. The little chip on each tank, which is intended to make life difficult for makers of third-party tanks, has been making my life difficult instead.
• Idiotic design. This is the killer. You may recall my concern that I would be left with a partially functioning device when one part failed. I need not have worried about that, because it seems the Canon design philosophy is one of extreme built-in obsolescence. When one part fails, even if it’s just an ink tank, the whole machine is a boat anchor. When the magenta ink tank is faulty, that doesn’t mean your prints come out looking rather less pink than they should. It doesn’t mean that you are restricted to monochrome prints. It doesn’t mean that you can’t do any printing at all. It means that the device is completely, absolutely, 100% useless. You may think that it should be possible to print in monochrome, scan a page, or send a fax without a cooperative magenta ink tank, but the Canon designers apparently think otherwise. What on earth were they thinking? I mean, how could anybody possibly think this is an appropriate design decision? Strewth!

Canon, this device is not good enough. I know that one person’s reliability experiences are not statistically significant, but even without that, the other downsides are enough to make me not want to repeat this unpleasant buying experience.

I have had very long life, 100% reliability and relative economy out of Hewlett-Packard devices in the past, so it looks like I’ll be returning to the fold with my next purchase. I know that HP doesn’t quite have the exalted quality reputation it once enjoyed, but it surely can’t be as bad as this. Can it?

That doesn’t mean we customers have to like it, and many of us don’t. Some of us happily accept it as part of the price of using our tools, some of us tolerate it with varying degrees of grace, and some of us fume about it but do it anyway because we have no choice. Beyond that point, there are those who won’t put up with it at all, and who find various ways around it.

Where a given user fits within that spectrum of views depends partly on that person’s personality, but also on the amount of inconvenience that the software throws up. I’ve started a new poll because I’m interested to discover what level of inconvenience people are prepared to put up with. What is a reasonable level of tolerance?

The poll question refers to a scenario where you have the need to use a piece of software, and before you can do so, you need to obtain codes. The question is not specific to AutoCAD or Autodesk, it applies to any software that needs magic numbers before it will work. Software often allows grace periods before a code is needed, but there are also circumstances where such grace does not apply, so for the sake of this poll please assume that you need the codes before you can get on with your job. How long would you be content to wait for the codes and still remain a satisfied customer?

Which of the following best describes your primary purpose for today’s visit?
. Other
To see if the discussion groups are still broken

How often have you visited Autodesk.com in the past 6 months?
. 6 times or more

A question about my industry group that didn’t want to copy and paste…
. Other
Question is not relevant

Do you currently own an Autodesk product?
. Yes

Are you planning to make a purchase decision related to an Autodesk product?
(I don’t know what choice to make here, none of them really fit. I’m on Subscription but that doesn’t mean I’m not involved in purchasing decisions; I am. I don’t know when the next purchasing decision will be, though. I picked:)
. No.

Which of the following titles best describes your role in your company?
. IT Manager

From which region are you accessing this site?
(Can’t you tell?)
. Australia / New Zealand / South Pacific

How would you rate your overall experience with Autodesk.com today?
. Very bad
(Actually, I don’t really know because because I haven’t yet got to the discussion groups I asked for, so I’m taking a wild guess based on recent experiences. I later checked the discussion groups and found that this was an accurate guess.)

Based on your best online experience, how would you rate www.Autodesk.com as a site that…

(Now, notice that is’s asking about my best online experience. I assume that would be best ever? Going back years, right? Before the recent update, then? OK, I’ll answer fairly based on that assumption.)

…is a reliable source of information that you trust?
9 Very Good

…leaves you feeling that your time was well spent?
8 Very Good

…helps you make well-informed decisions?
9 Very Good

…is easy for you to navigate?
5 Fair

…allows you to move rapidly to the information you need?
6 Good

…enables you to find what you’re looking for?
7 Good

…encourages you to return?
7 Good

…meets or exceeds your expectations?
6 Good

…you would refer to others?
7 Good

…has content that is relevant to the purpose of your visit?
6 Good

…gives you the amount of detail you need?
6 Good

…covers the range of information you need?
7 Good

…enables you to identify and contact the right people?
0 Very bad

…provides a positive interactive experience?
8 Very Good

…enables you to help yourself?
7 Very Good

That’s the end of that section, the progress bar is half-way though, so I go to the next section, which I assume is going to ask the same questions based on my worst experience. Oops, no it’s not! The survey is over! Thanks for playing.

Now you know. So, if in a few weeks somebody from Autodesk refers to “survey results” that supposedly show how well the recent update went down with users, point them at this post. I gave high marks for some of my responses, but I wasn’t being asked about my experiences after the recent update. I was being asked about my best experiences, which is altogether different.

This sort of thing is why I never take survey results from anyone at face value. I always insist on seeing the full details, otherwise I will give such results no credit at all. No details, no point.

The editor will cunningly allow you to apparently fix up these errors in the places they occur, and then the fun-loving little sprite will reintroduce the same or similar errors as soon as you save the changes. Multiple edit attempts will get you nowhere (except a padded cell, perhaps). Somebody must have had a wonderful time writing that one.

Another bug relates to the display of quoted messages. Admittedly, this was always going to be a difficult task to get right in the new environment, because of the many quoting styles that exist in the messagebase. No surprise, then, to discover that quoted text frequently displays in such a way that makes the message author look like a clueless dolt.

In related news, I’ve added a poll that asks what you think of the recent web update. I’m not making my usual attempt to remain neutral and avoid influencing the poll results this time, as it’s a bit late for that. Everybody knows my views by now, but I suspect it wouldn’t make much difference in any case. People are angry enough about this mess without any influence from me. However, it’s always good to see a wide range of views expressed; somebody thinks the update is “Fantastic”.

However bad the discussion groups are, at least the Subscription site is working (for me anyway, I know there are still people with login ID problems) and my helpful Indian chappie came back to me with a workaround. It’s not a very good workaround, and it only applies to Subscription customers, but I thought I would pass it on anyway:

Log in to the Subscription Center, pick Search in the top right corner, then fill in your search details or pick Advanced Search for more control.

This search method does find messages that date back before the recent web update. However, there are a few problems with it. There’s no way to restrict it to just discussion groups. Even if I restrict it to just “Communities”, it returns results that include various blogs, and to threads that have been moved or deleted. If more than one page of results is found, there’s no way of going directly to a given page, it’s Next > Next > Next > Next > repeatedly. If I try to restrict the search to AutoCAD 2009, for example, it returns nothing. Finally, it’s obviously only any good for Subscription users.

Another workaround is to use Google Advanced Search and set the Search within a site or domain field to discussion.autodesk.com. However, I know of no way of restricting the search to AutoCAD 2009, for example.

Enough band aids, the Autodesk discussion group search mechanism really needs fixing, along with all the other problems. I’ve already seen suggestions that Autodesk sabotaged its discussion groups on purpose. Personally, I’m generous enough to think that it’s just gross incompetence, but Autodesk’s continued silence and apparent inactivity can only encourage the conspiracy theories. I don’t know how much Autodesk pays for PR each year, but I bet the negative impression from this disaster is worth a lot more than it would have cost to have just done the job properly in the first place.