Interview with Fink's Project Leader

Gentu writes "There is a interesting interview over at OSNews with Fink's project leader, Max Horn. They discuss Fink's relationship with Apple, integration of their Unix/Linux ports to Mac OS X via Debian's packaging solution, ease of use on installation of the .deb packages, AltiVec optimizations and more."

Comments

Re:Fink is a problem

[ProfKyne]

Instead of trying to create software that piggybacks on GNU/OS X why aren't they working on GNU/Darwin?

You're missing the whole purpose of Fink. It's a porting project. Of course there's coding work involved, but the idea is to get software that runs on other unices to work on OS X / Apple Darwin.

Re:Fink is a problem

[Llywelyn]

Yum, a troll.

"They can't seem to get that NIH attitude out of their heads even when they go open source."

You mean like USB, Ultra ATA, and Gigabit Ethernet? Yeah! Apple needs to be more standards compliant damn it and support... oh, they *do* support them and have for quite some time.

Or do you mean things like gcc3.1, gprof, XFree86, libtool, automake, KHTML, &c?

"Fink and GNU/Darwin are clean up efforts in reponse to Apple's failure to:"

You do realize that it isn't Apple's responsability to do everything for you?

Boo hoo! it didn't ship with TeTeX, how am I ever supposed to do my document typesetting! They must not be serious about the Unix community or they would include such essentials that of course 90+% of their userbase--both users and developers--is going to use.

" X11 from day one. "

I don't suppose that you have heard of the XonX project?

"useable AND would have convinced more traditional workstation users that Apple is serious about open environments."

This statement does not grok. Most "tradiational workstation users" don't give a damn whether Apple is "serious about open environments."

"The official X11 for OS X package remains in beta to this day."

Considering it was released in January and we are on Beta 3 right now, which is more than useable. It took Mozilla how long to get through its own Beta stage?

"Fully embrace a mainstream open source OS. "

'They have done so much that no-one else in the field ever has... but damn them! They aren't doing enough!'

Am I getting the gist of your argument?

You do realize that, for the vast majority of us, we try to use what works best for us and do not choose our platform soley on the basis of religious zeal.

" I don't care how black-turtle-kneck-sweater cool the MACH kernel is"

The variant of mach they are using is almost, but not quite, entirely unlike mach. It has been so heavily revised it isn't even funny.

"If they aren't ready with a mechanism for weekly or daily patches,"

You mean like CVS?

Incidentally, so Apple left something for groups to do--such as produce fink--what is your point?

Also, am I the only one who doesn't install the newest library just because its available? Seriously, you do realize I have software that I need to have work?

Re:Fink is a problem

[Llywelyn]

"Instead of trying to create software that piggybacks on GNU/OS X why aren't they working on GNU/Darwin?"

Maybe because there isn't such a thing as GNU/OS X?

Less flippantly (but only slightly): Maybe because the purpose of fink is to improve the ease of UNIX integration with Apple's MacOS X?

"It seems to me that they don't like apple very much for some reason."

You must not have read the same article I did.

The one that said:
<<Overall, the mood at Apple seems to be friendly towards Fink, they refer to us in various places of their homepage, for example.>>

" Political reasons maybe?"

No, it would be for "political reasons" if they suddenly started working working exclusively on the GNU-Darwin project.

"Maybe they should have bought PCs instead of a MAC and run linux."

You do realize that it isn't an acronym? Mac, short for Macintosh.

You also realize that most of us use OSX because we happen to like OSX and because it works for us? *Not* for solely political reasons?

Re:Fink is a problem

[Halo1]

(note: this is a reply to both this and your previous post)

Do a search for the bugs Apple has introduced to IPSec and GCC.

And you do a search on the improvements they introduced in those things. Yes, when you do improvements or change things to improve compatibility with your product, you always run the risk of breaking things. Regression testing can only help you so far.

Why can't Apple's out-of-the-box Unix be as compelling to RedHat users as say

Maybe for the same reason the the reverse also doesn't necessarily hold true. You are now already complaining that Apple ships some buggy stuff. Do you think the situation would improve if they started shipping a lot more programs? Red Hat also didn't create/port all that software themselves, nor are all the maintainers Red Hat employees. Linux is simply a more established/older Unix variant, so more software has already been ported to it and thoroughly tested.

Perhaps I've failed to make my point that in my humble opinion, Mac OS X 1.0 should have shipped with an X11 implementation right out of the box given that is shipping with say... developer tools.

What do developer tools have to do with X11? You can perfectly create programs that have nothing to do with X11. In fact, a non-X11 program is likely to be of much greater value (in the sense of usability and attractiveness) to the Mac community than an X11 program.

It's true that Mac OS X is probably the only modern Unix variant that doesn't ship with X11 by default (yet), but has it ever occurred to you that the reason for that may be that Mac OS X already has a different window manager on board (while for the others it's either XWindows or nothing) which is much more important to most Mac users? I really don't understand why you are so obsessed with X11 not being included by default.

Fully embrace a mainstream open source OS

Those "mainstream open source OS"'s lacked several key features Apple needed in its OS at the time Mac OS X development started and they are only now catching up. Which features? For example proper smp support (multi-processor Macs finally had to be able to their fullest potential), proper real-time support (Final Cut Pro anyone?) and low latency support (low-level audio, user interface).

I really don't understand why you think Apple should do everything based on open source. What is the problem with using existing stuff of which you think it's done well, and do other things (of which you think you can do them better) yourself? What is the problem with first focussing on your largest user base (Mac users) and only when you're confident you're on the right track there, starting to spend resources on other target groups?

I'm not trying to be elitist here, I'm just trying to point out that Apple is a company with a small market share and thus it can't take very large gambles (it doesn't control the market). You call opening the Darwin source (and several other packages) a leap of faith, but there's no large risk involved there. At worst, it has cost them a bit extra in terms of bandwidth and HD space. On the other hand, it gave/gives them extra publicity, it makes it much easier for driver developers and they get some free bug fixing in the process. On top of that, the open source world also benefits from it, so it's a win-win (or WIN-win for Apple according to a lot of people, but even then everyone still gains) situation.