Update-itis
Jun. 29th, 2014 08:48 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
blatherskiteI've been using Adobe software for scary long... since 1987, if memory serves. Back then, Adobe had a reputation for quality. PageMaker, for example, was rock solid: it never crashed, and it produced decent type if you knew how to use it. The feature set was limited compared to modern desktop publishing software, but everything was there that you needed to produce nice books -- and I produced a great many over the years. Updates came along at reasonable intervals, and either brought important new features, or at least fixed longstanding bugs. That stopped somewhere around PageMaker 5, around 1993, at which point Adobe fell into the trap of upgrade-itis.
Upgrade-itis is a disease I first encounted with Microsoft, in the form of the autonumbering bug that existed for more than a decade before (maybe) being fixed around the time of Word 2007. The diagnostic symptom of upgrade-itis is that you stop worrying about old bugs that people have learned to live with or work around, or have at least stopped complaining about, and instead introduce new features on an annual schedule, whether or not the software is ready and whether or not the features actually meet a need. The goal is to ensure a steady flow of income, not to make your customers happier or more productive. A related symptom is that you stop optimizing both old and new features, because you don't have time; creating new features becomes the obsession.
One Adobe example is Dreamweaver. Going back several (Mac) versions -- since about 2006 -- there's been a bug related to basic cursor movement. When you formatted a chunk of text as one of the built-in heading styles, the text suddenly drops out of the regular flow of text that you can clearly see on the screen. So, for example, if you place the cursor on the line above the new heading and press the down arrow key... nothing happens. You can't scroll past the heading. Click to position the cursor inside the heading, and now you can't use the arrow keys to scroll upwards or downwards out of the heading; you have to reach for the mouse. Basic commands such as Command+PageDown that should take you to the end of the file sometimes stop working for no obvious reason, even though PageDown continues to work. You just have to press it many times to get to the end of a long file. There's also the inconvenient problem that periodically arises in which Dreamweaver forgets the modification dates of your files, so that it feels obliged to reupload your entire Web site because it believes the versions on your Web server are out of date. (To be clear, this is not an issue of updating library content so that all files that include that content are updated. I've confirmed that this isn't the case.) When I was keynote speaker at STC India's conference a few years ago, I had an opportunity to demonstrate and document these problems for several key Adobe staff. The problems remain unsolved 8 years later. As they've persisted across versions and several new computers with new operating systems, I don't think the problem is limited to me.
InDesign has its own share of problems, in part because they adopted seriously problematic code for the indexing feature from PageMaker, where the feature never worked right. Index entries that span multiple pages never worked reliably in PageMaker, and 6 versions of InDesign later (I'm currently using the CS6 version), the problem is still there. Worse yet, despite the frenzy to get InDesign and its companion applications up into "the cloud", nobody's bothered to actually look at how indexers create indexes. So index entries are bound to the chapter in which you create them; you can't reuse index entries from a previous or subsequent chapter unless that chapter is open for editing (which is problematic for a book such as my current project, with 24 chapters). This leads to an enormous amount of unnecessary retyping and many consistency errors that take considerable time to fix when you proofread your index. But hey, if I upgrade to the new "creative cloud" version, I can at least save my files to the cloud... on those days Adobe's cloud services aren't down or hacked.
The underlying theory seems to be that maintenance upgrades that fix problems and create a stable product instead of introducing cool new features don't generate income. And to be fair, companies as large as Adobe require a large and ongoing income stream to support their activities. But just once, I'd like to see a company stop, figure out what's wrong with their software, and just release an entire new version with no new features... but that fixes all the old broken features. (I have a vague memory of Microsoft doing this several years ago, but the specific product name isn't coming to mind... possibly it was a version of Windows XP with a batch of problems from previous versions finally fixed, which Microsoft shipped to anyone, free of charge, on a DVD.) Customer loyalty counts for something, and such a gesture, while it might temporarily decrease the company's stock price, is likely to restore shaky loyalty caused by a long series of problematic feature-laden releases. In the long term, that's a recipe for corporate health, not a lost opportunity to earn more money in the short term while giving customers a strong incentive to look elsewhere for a better product.
Upgrade-itis is a disease I first encounted with Microsoft, in the form of the autonumbering bug that existed for more than a decade before (maybe) being fixed around the time of Word 2007. The diagnostic symptom of upgrade-itis is that you stop worrying about old bugs that people have learned to live with or work around, or have at least stopped complaining about, and instead introduce new features on an annual schedule, whether or not the software is ready and whether or not the features actually meet a need. The goal is to ensure a steady flow of income, not to make your customers happier or more productive. A related symptom is that you stop optimizing both old and new features, because you don't have time; creating new features becomes the obsession.
One Adobe example is Dreamweaver. Going back several (Mac) versions -- since about 2006 -- there's been a bug related to basic cursor movement. When you formatted a chunk of text as one of the built-in heading styles, the text suddenly drops out of the regular flow of text that you can clearly see on the screen. So, for example, if you place the cursor on the line above the new heading and press the down arrow key... nothing happens. You can't scroll past the heading. Click to position the cursor inside the heading, and now you can't use the arrow keys to scroll upwards or downwards out of the heading; you have to reach for the mouse. Basic commands such as Command+PageDown that should take you to the end of the file sometimes stop working for no obvious reason, even though PageDown continues to work. You just have to press it many times to get to the end of a long file. There's also the inconvenient problem that periodically arises in which Dreamweaver forgets the modification dates of your files, so that it feels obliged to reupload your entire Web site because it believes the versions on your Web server are out of date. (To be clear, this is not an issue of updating library content so that all files that include that content are updated. I've confirmed that this isn't the case.) When I was keynote speaker at STC India's conference a few years ago, I had an opportunity to demonstrate and document these problems for several key Adobe staff. The problems remain unsolved 8 years later. As they've persisted across versions and several new computers with new operating systems, I don't think the problem is limited to me.
InDesign has its own share of problems, in part because they adopted seriously problematic code for the indexing feature from PageMaker, where the feature never worked right. Index entries that span multiple pages never worked reliably in PageMaker, and 6 versions of InDesign later (I'm currently using the CS6 version), the problem is still there. Worse yet, despite the frenzy to get InDesign and its companion applications up into "the cloud", nobody's bothered to actually look at how indexers create indexes. So index entries are bound to the chapter in which you create them; you can't reuse index entries from a previous or subsequent chapter unless that chapter is open for editing (which is problematic for a book such as my current project, with 24 chapters). This leads to an enormous amount of unnecessary retyping and many consistency errors that take considerable time to fix when you proofread your index. But hey, if I upgrade to the new "creative cloud" version, I can at least save my files to the cloud... on those days Adobe's cloud services aren't down or hacked.
The underlying theory seems to be that maintenance upgrades that fix problems and create a stable product instead of introducing cool new features don't generate income. And to be fair, companies as large as Adobe require a large and ongoing income stream to support their activities. But just once, I'd like to see a company stop, figure out what's wrong with their software, and just release an entire new version with no new features... but that fixes all the old broken features. (I have a vague memory of Microsoft doing this several years ago, but the specific product name isn't coming to mind... possibly it was a version of Windows XP with a batch of problems from previous versions finally fixed, which Microsoft shipped to anyone, free of charge, on a DVD.) Customer loyalty counts for something, and such a gesture, while it might temporarily decrease the company's stock price, is likely to restore shaky loyalty caused by a long series of problematic feature-laden releases. In the long term, that's a recipe for corporate health, not a lost opportunity to earn more money in the short term while giving customers a strong incentive to look elsewhere for a better product.