Apple and OSS (Re: Why reinvent the wheel?)

by Matthew Russell

In my recent article How Does Open Source Software Stack Up on the Mac?, a reader and I have been having a very interesting discussion via the talkbacks at the end of the article that has drifted through various facets of OSS, economics, and ethics. I'd like to invite you to read the Why reinvent the wheel? discussion and chime in here with your thoughts. An opinionated summary of the most interesting thoughts that have come up so far follows, along with my take on each of them:

7 Comments

Dogzilla
2006-07-30 21:19:30
Hey there. So to continue the conversation (thanks for giving this discussion its own forum, btw), I'll add my own quick thoughts:

  • I don't necessarily believe that OSS per se drives Apple to innovate. I think good software of any kind does. But overall, I think innovation is an integral part of Apple's DNA. Even in the utter absence of any competition, my belief is that Apple (the current Apple anyway) would be driven to innovate in any market they identify as core to their business. In other words: Digital Hub apps = resources allocated to innovation; Server applications: not so much.

  • Let me first say that I absolutely agree with you on your "stealing ideas" stance. I've always likened it to prose: stealing ideas makes you a great artist, whereas stealing actual words makes you a plagiarist and/or pariah. Having said that, I still think Apple has to be careful and concerned about generating buzz that they're "stealing ideas" or "crushing small developers". The Dashboard/Konfabulator generated some amount of bad feeling towards Apple from small developers, reagrdless of it being justified or not (my own opinion is that Apple was entirely within the bounds of propriety - Konfabulator wasn't the first, unique implementation of the concept. It seemed to me to be a trimmed down version of the OpenDoc idea more than anything else) I think that every feature that gets added to a bundled Apple app which overlaps the features of existing 3rd-party software has to generate some calculus within Apple of desirability vs. ill-feeling. Even if - as you point out and I agree - tabs are a "commodity concept", there's still a small developer that added them to iChat.

  • I don't agree that having a large number of applications with exactly the same feature set does anything to improve the marketplace. It could be argued that confusion in the marketplace could be a detriment. I would posit that having a small number of quality apps offering similar core functionality and unique extended feature sets is a preferable situation and a sign of a healthier market. I would choose a marketplace with only TextWrangler/TextMate/SubEthaEdit over one with 25 Java Emacs clones every time.

  • Disclosure: I actually am an Apple shareholder. That being said, I'm not sure about this one. As I alluded to before, I don't think there's a clear answer on this one. My gut tells me that this is like porn: you know it when you see it. In some cases, Apple adding a feature or entering a marketplace just makes sense, either because they feel they can do it better, because the app is strategically important and they don't trust the 3rd-party vendor (*cough* Adobe *cough*), or because the product is a natural choice to incorporate into the core OS (eg: Sherlock 3/Watson). But I still think Apple has to keep a wary eye on their developer relations, especially on the small ones. At least, as an Apple stockholder, I hope they do, as this seems to be a real long-term growth strategy.


I also wanted to pose a question of my own: how much do you think Mac users differentiate between OSS software and shareware / commercial products? For many OSS advocates, using OSS seems to go beyond mere quality / feature-set comparisons to a point of pride and/or religious fervor (eg: Cory Doctorow). But it seems that the situation with the Mac has blurred the clear-cut distinctions that existed before, where it seems more likely to find commerical products coexisting with OSS software. To what extent do you think this is true, and if it is true, what does it mean for the quality of and expectations of OSS software?
ptwobrussell
2006-07-31 05:06:58
@dogzilla:


how much do you think Mac users differentiate between OSS software and shareware / commercial products?


Distinguishing between shareware and commercial products on a Mac can be tricky business too I think -- much trickier than on Windows -- and I think that too bears testament to the way I'd personally respond to this thought. Apple seems to have have made a highly concerted effort to allow small developers to produce amazing applications by opening up the best APIs imaginable and providing many of the same developer tools that their own engineers use. In that regard, it's easy to see how they really try to foster this kind of development, and that too goes a very long way toward generating non-commercial high quality apps. Let's face it, without the will/know how from developers we wouldn't be having this discussion.


Because of the abundance of options to choose from, including apps that Fink and DarwinPorts have made available with very little overhead involved for the end user, I really do think the lines are quite blurry. Now, having come to the point of explaining the atmosphere a bit, I think it helps us to understand some things about Mac users: we appreciate and support small developers, we've come to understand that you don't have to be a corporate giant to product innovative, high-quality software, and finally, the many options that the underlying BSD core have brought to the table from the linux community can really be useful -- often times even more useful when they're running on a Mac than when they're running on a linux box. In general, that means that we have lots of choices, ranging from "joe's latest project he's giving away for free" to commercially supported. I say all of that just because I think it's pivotal to understanding the landscape.


Since I'm a developer by trade myself, I tend to favor OSS whenever possible, because I appreciate the "freedom" (as in source code, not price) that it gives me to hack it if I want/need to, and of course, you can't complain about the price. I guess it also goes without saying that I don't mind sinking time sometimes with a product that's half-baked. I think the whole ideal of "community involvement" goes a long way too. I always have thought of OSS as a powerful form of community service, and I want to give the developers incentive to keep up their good work by supporting what they do monetarily, in the forms of downloads, buzz, etc. In the end, I belive this creates a win for everyone.


Certainly, the quantity of high caliber developers along with the rich API and tools available on OS X raises expectations of OSS and other forms of shareware/donationware/freeware/etc. and that's a great thing since it raises the signal/noise ratio for OSS. Given the diverse options available for OSS, the concepts involved in the various theories of darwinism seem to especially apply here, since lower quality OSS is especially prone to being ignored or overlooked in a landscape where certain established products are "higher on the food chain."


So to sum that all up: I think it's become more and more difficult to clearly distinguish between commercial and non-commercial products. In a way, Macs are increasingly becoming melting pots for the software community, and this no doubt brings new ideas, creativity, and innovations to the table. Fortunately, much of this good stuff is coming from the OSS community. Sure, there are fanatics out there who would choose to use a half-baked OSS product instead of purchasing a higher quality commercial product "just because" but I admire what I perceive those folks' motives to be: harboring a higher quality OSS landscape. And you can't really do that if you just blindly ditch a product until you think it's come around without trying to help it come around in some form or fashion.

Dogzilla
2006-07-31 06:55:19

Hmmm. Well, I wholeheartedly agree on the point concerning OSS software on the Mac - I feel there's little differentiation in a consumer's mind as to whether an app is OSS, commercial, or shareware. Apps basically get judged on their own merits more than on whether they're open source or not.


I think this is a two-edged sword: while it raises visibility for OSS generally, there are many, many OSS projects that simply aren't very good. And even where they are useful, they lack polish. In some ways, I think getting dumped into the larger bucket of "Software" will be good for OSS projects: you can get away with a lot when your main audience is other geeks, but when average Joes start using your software, you need to aim for a higher level and dremel off some of those rough edges. I'd say with Mac users, even more so. If OSS software is to attain the critical mass that so many OSS proponents hope for, it needs to go through this process.


As for "blindly ditching" a product - I don't think I'm doing so at all. I simply don't find any of the OSS offerings in the "iApp" sector to be anywhere near as compelling as the Apple offerings. While I freely admit that I'll ocassionaly come across a feature that makes me say "Oh, cool", for me that doesn't translate into "I'll move everything into this new app". I guess I'm past the stage of sinking time into an app just because it's open source, especially for apps that aren't core tools for me. On the other hand, I am an active member of the communities for those tools that I use every day (FreeBSD, PHP, various commercial apps). Point is, there are a *lot* of people - especially in this new, larger market into which OSS is entering - that don't really care about the (for lack of a better word) political aspect of OSS. They simply care whether it works or not. To bring the discussion full circle, those folks are the hearts and minds that OSS needs to capture now, and highlighting the uniqueness of OSS plays to its strengths, whereas discussing me-too apps will generally leave the aforementioned software consumer lukewarm. Showcasing apps like Scribus, NeoOffice, and Seashore (I realize you pointed out some of these, but they received what felt like a passing mention) would be more likely to excite and involve the average software consumer with OSS than, say, AlbumShaper ("I can use Scribus for free instead of paying a $400 Quark Tax every 2 years?!").


I hope this makes sense. I honestly feel we're standing pretty close together in our viewpoints, and that we're not really on opposing sides in this discussion. Chalk it up to me being an old fart, but I think the differences are more due to things you accept as givens which I don't. I strongly support the concept of OSS, but I won't give it unquestioning loyalty - I think it has to prove itself just like any other piece of software. Ultimately, I think this will make OSS stronger.


2006-07-31 17:24:58
I think that Apple's own apps tend to err on the side of simplicity - for instance, tabs are optional on Safari. There is a great quote on Joel On Software about how Microsoft engineers have 'Never seen a feature they don't like' whereas you feel that there is a lot of argument about features added into the main OS X apps. That isn't meant as a criticism in any way - some people have always dismissed Macs as 'computers with stabilisers' as if there was some inherent value in things being difficult to use or complex.


That means, however, that there is plenty of room for applications that offer a more powerful feature set for more 'advanced' users - whether they're OSS or commercial.


Number of apps : I whole-heartedly agree that having a small number of quality apps is better. One of the things I like about the Mac is the fact that I can rapidly find a good application for a specific need. WebKit is an interesting example of where developers are making interesting extensions of a core piece of functionality - from the instant preview in TextMate through to browser's like Shiira and OmniWeb starting to move beyond Safari 2.


I can well understand why some people are politically committed to OSS, and for those people having OSS apps that re-invent the wheel is important.
In fact, in some ways I actually agree more with RMS point of view - that users of FOSS have to accept that it may lag behind proprietary software in features but that this is the 'price' of freedom.
(He presents this in contrast to the business friendly argument that OSS results in better, more secure and bug-free applications)


Of course he also suggests that users of said software should pay to have the missing features developed, but no one seems to have really been able to make a good model out of that yet - the only paid FOSS development seems to be in the enterprise world (as a loss-leader for service contracts) as opposed to application development.


I think that concentration on the service-funded model is really holding FOSS back. Companies are happy to pay service contracts but home users prefer to make one off payments (even if they are annual a la iLife).


The other thing is that FOSS advocates often only see the feature set rather than realising that a lot of the value in apps like iLife or iWork is in the templates. Apple employ a lot of non-programming staff on it's applications - whereas very few FOSS projects do.


Cory Doctorow actually puzzles me in that I didn't know that he wasn't already a Linux user - I'd have thought someone so involved in the EFF would have switched years ago!!


Overall, however, I agree with the view that OSS is beginning to drive innovation at Apple. For a long time Linux has been playing catch-up with Windows, but the focus on GNOME and usability in distributions like Ubuntu and Novell's SLED/SUSE, is beginning to bring it closer to OS X.


In some cases - like multiple desktop support - it is even ahead. I think that's another case of Apple ignoring something a small minority of techies and power-users want in favour of simplicity (I've played with Virtue and it just doesn't seem to work comfortably for me with the Dock. Maybe if I ditched the Dock wholly in favour of Quicksilver . . .).

eon
2006-08-01 14:54:32
Didn't Firefox come after Safari was release so that Mozilla could focus on creating something less bloated than Netscape?
keath
2006-08-07 08:48:24
Why reinvent the wheel?


The only reason is when you aren't satisfied with what's available, or with the cost. Apple applications are still free with your Mac purchase, or for a fairly reasonable upgrade price. Most people have no complaint there.


The applications are generally reviewed as excellent. Not easily topped in ease of use, and with enough power for most users. There just not enough complaints overall to warrant replacing them. Of course nothing is perfect for all users, and power users want more. There is opportunity for developers to make more powerful versions of Mail, iCal, and the other iApps to satisfy others needs (although Apple's professional line has some of that pretty well covered).


As for OSS vs. commercial developement, it makes no difference to me. I happily pay for good software that I will use, even where free alternatives exist.


The problem with many OSS projects is that as cross-platform software, the developers never focus on Mac GUI design. As a user I'm just never going to be satisfied with that.


The best of open source for me is Apache, PostgreSQL and perl (insert your own language or database here). These are projects that add a great deal to the Mac platform for both users and developers. They are software projects that are considered best in class, are freely available and easily compatible due to the Unix base of OS X; so Apple has no need to try to provide a solution of their own. Again, even though commercial databases and webservers are available, most users will be more than satisfied considering the price/performance of what is freely available. Non-GUI server side stuff is one area where community software has really done well.

Dogzilla
2006-08-09 08:36:17
Maybe this discussion is a little dated now, but I thought this might be relevant to some of the points that came up here.


7 Apps on Leopard's Hit List