Chad Perrin wrote:

> It appeared pretty clear to me that Robert's point was that, like all
> proprietary, closed source software, TextMate may one day disappear from
> the market simply on the whim of the copyright holder (or because the
> copyright holder "goes out of business", gets hit by a bus, whatever).
> As such, putting all your eggs in the TextMate basket may be kind of a
> losing proposition.

Well ... I'll have to go look at Robert's post, but I don't think open
source vs. closed source is a significant factor in the viability or
longevity of a piece of software. I personally think MacOS is a fine
product, and I think the same thing about Windows XP Professional SP 2.
And I think the same thing about the major breeds of Linux I've dealt
with: Red Hat/CentOS/Fedora, Debian/Ubuntu and Gentoo.

> There are other reasons to prefer open source over closed source, as
> well.  All in all, I don't think TextMate is good enough that it
> overcomes the negatives of being a closed source, proprietary
> application.

Neither do I -- requiring a Macintosh is a deal breaker. I'm not against
buying software. If you recall I compared KDevelop and Komodo a few
months back in my search for an industrial strength IDE and ended up
purchasing Komodo. Why?

1. Komodo was cross-platform. Not only is KDevelop Linux-only, it's
really KDE-only and Qt-only. Linux, Qt and KDE are also fine products,
but I didn't want to limit myself to them.

2. Komodo did a better job of syncing with my RubyForge repositories
than KDevelop did.

3. I didn't need C/C++, which KDevelop supports and Komodo doesn't.

> There are cases where, in a strictly technical sense, the
> benefits of a given piece of software *do* overcome the detriments, but
> for my money this is not one of them.  This in no way means that whether
> or not something is open source is the only, or even biggest, concern,
> but rather that whether it's open source is simply an *important*
> concern.

To me, viability and energy of the development team are more important
than whether I can hack on the source or not, except for software in
very specialized areas. For example, there are a lot of Petri net
analysis and modeling packages out there that are freely available to
the academic community in source form but not for commercial users.

I'd like to hack on such stuff, but I can't. But I *don't* want to hack
on an IDE, or a word processor, or a browser. In those cases, open
source only means I probably don't have to pay for it and I return the
favor by filing bug reports rather than trashing them in my blog. * :) I
want IDEs and browsers and word processors that do my bidding and make
easier hacking on the stuff I want to hack on. :)


* Except for Firefox and OpenOffice -- those I *have* trashed in my blog. :)