Allan Odgaard made it plain in a recent post that TextMate 2.0 will require Leopard.
I thought his list of “costs” for maintaining backwards-compatible code was very interesting:
- Time spent debugging (and sometimes making workarounds for) issues only present on the older OS version.
- Time spent implementing stuff that Apple offers for free on the new version of the OS.
- Not being able to make use of features only present on latest version of the OS when it’s too impractical to conditionally make use of them.
- Code complexity, because it needs to do different things on different versions of the OS.
The vast majority of the comments on Allan’s post are supportive. What’s most telling of all is that 2.0 will be a free upgrade - a hugely impressive offer, considering how cheap TextMate is to buy in the first place.
I’d imagine there are a number of other small software companies and one-man-bands who are in the same position as Allan. How about you? Will your app require OS X 10.5?
No. Limiting your software to a particular platform only makes a small amount of sense ( if any ), limiting it to a very limited flavour on that platform is odd to say the least, but hey it's up to him. And it depends what it is of course. But where I work we strive to make software that works and looks good accross platforms ( PC and Mac ) and to span as wide a version gap as possible and it pays off too !
Leopard seems very pretty to me, but actually it offers very little in terms of "new" "must have" "must use" services so what is it exactly that a text editor needs that limits it to such a small market ? His excuses seem fairly lame to me. 10% ? I wonder how he came up with that figure.
I guess he isn't doing it for his main income!
It's his call, but I think it makes very little sense. Also with almost 20 years of Mac programming under my belt I've learned it's nearly always better to NOT to adopt too many new Apple ideas in one gulp until they've at least gained a bit acceptance. ( OpenDoc anyone ? ) There are always exceptions to this rule though !
We simply cannot afford to do that. In our case, we have a product that runs on many different operating systems, and end-users expect a certain consistency across the platforms. Additional costs to consider in our case are the impression it would give to our customers (even the non-Mac ones), and the time our support people may have to use up explaining to people why this is the case. (This is not to mention the time that the developers will have to spend explaining these things to the salespeople, who may have already promised to support an older version of the OS. :^)
The app I'm currently working on is going to be for Leopard only. Some of the features of 10.5 are just too attractive to pass up. CoreAnimation and properties in objc2 at the top of the list for me.
ElRic: Allan does do it for his main income. And I'm pretty certain his numbers are good.
For myself, 97% of my users are running 10.4. And when 10.4 originally came out, the majority of my users upgraded to it within weeks. It just doesn't make sense to support older OS releases, especially for new versions of an app.
Hi
I don't do app programming, but it seems that Apple knows how to boost OS transitions : adding nice features dedicated to the developers in each release, which encourages them to make nice features in their software, using the shiny new components, which will make users upgrade faster. For example Textmate 2.0 is certainly one of these app that will make me switch to Leopard.
I am taking a very similar approach for Bee Docs' Timeline. As a solo developer working part-time on the project, there has to be a compromise. I can either build a full featured application that is well tested and delivered in a reasonable time (but is limited to one platform). Or, I can strip down the features, and have much longer development and test cycles (and support a wider range of platforms).
There are two other important factors for me. The first is that there is already a very nice version of my software which works on older systems. This is the situation with TextMate as well. It isn't that people with older versions of the OS will be left in the dark. They just won't have the latest and greatest features. Most people who thirst for the latest and greatest features from their favorite applications, are also going to want the latest and greatest from Apple and will upgrade to Leopard.
Last but not least, much of my motivation for building software as an independent developer is the joy of designing and building cool software. Finances are only one factor, and not the most important one. I have the most fun when I am able to take advantage of all the latest features of Mac OS X and focus more time on design and implementation rather than testing and support. I think the sense of "going with your heart" is what makes much of the independent Mac software so much fun for customers to use.
At the end of the day, he knows his market. It's not a consumer application, but a developers editor - and it doesn't have a large legacy user base like BBEdit - yet.
Time Machine suggests some sort of file versioning system built into the OS. I can see that being something very useful to have integrated into a text editor. The system wide To Do/iCal APIs are also a feature I am looking forward to various applications exploiting. Probably the most likely candidate, however, is the one with no obvious visual feature - Obj-C 2.0.
It's obviously his call, but in many ways it would disappoint me and I'm a big fan of TM and a paying customer. If it's a free upgrade obviously I'll grab it as soon as I can and have Leopard installed, which is as soon as possible on a test machine at least. But I'm unlikely to run Leopard on my production machines for quite some time after it's released which then means I can't take advantage of the new TM. For me as a user it doesn't matter whether what he implements requires some Apple things in Leopard or not. All that matters is the quality of the software and the new features. And he forces me to upgrade the OS to get the new features. But I can't blame him much really. He wants to keep TM on the cutting edge.
I'm not developing professionally, but it seems to me that Allan's clientele is going to upgrade to Leopard pretty much immediately (I know I will). As he does all the work himself I can see how it would make more sense for him to save the development time.
Also I have heard it said, although I have no first hand knowlege, that Leopard will be an upgrade the developers will be very interested in pushing consumers towards as quickly as possible. While 10.5 has not dazzled any consumers (yet) and lacks a killer feature like Expose or Spotlight (yet), it seems to me that it has a lot of under the hood improvements that developers will want to take advantage of as quickly as possible.
One factor that I've heard over the years, from small software developers, is that people with the latest version of the OS are more likely to spend money on software. The perspective, in brief, is that you're keeping up with Apple you've got more money, more interest in your computer, and being attentive to the market. It's like the famous 18-35 year old demographic--the sweet spot of marketing.
Could it be that, as a text editor, TextMate will be utilising some of the new text underpinnings in Leopard? Just a thought :-/