Free Lossless Image Format (FLIF)


Hi all :)

@ _wb_ : how about submitting your work to Xiph.org ?

On a side note, the people on that PNG mailing-list seem very nice and knowledgeable, that's a good thing.

Cheers, Magic Sam
 
History is being written here and we can see it happening live, how awesome is that?

Here's something grainy for your animation tests:
tumblr_n79mxdrFKv1s2wio8o5_400.gif
 
I have to agree with the sentiment on the usage of GPL v3. This basically implies the format won't see any notable adoption.

It shouldn't, the license is only applying to the code for copyright purposes, not the algorithm.  If someone wants to use FLIF they have to re-implement the algorithm but it can be done.  It's a bit more of a barrier than just using the existing code but shouldn't be a showstopper.
 
Heh - flif.info was just shared on one of the bigger swedish-language FB groups :)
 
The number of people harping on the GPL thing is astounding.  Or perhaps it isn't, really; perhaps I have too much faith against kneejerk reactions.
 
I think we can all agree that a libflif under a license that's at least as permissive as LGPL is going to be necessary for adoption to take off at any meaningful level. Someone else can do it but starting evangelizing this way is going to be yet another major obstacle.

Pointing this out isn't much of a kneejerk or overzealous reaction IMO.

However, I do agree that there's a good reason to keep it GPL right now while the spec is not even finalized. This prevents people from forking it and making their own closed specification, although LGPL would also do that. But at this stage in development discouraging people from linking to it may actually be a good thing.
 
Last edited by a moderator:
I'd agree except that people aren't wording it like that, most are claiming it IS dead because it IS GPL3; complete dismissal.  That's the kneejerk I'm talking about, no consideration that it can be relicensed or reimplemented, just a dead fish before it's done.

On another note, I do like the idea of trolling everyone by pronouncing it "fly-f" :p
 
Its instinctual group-animal behaviour, you assume someone has a point when they exhibit a opinion, rather than questioning the rationality behind doing so.

The reason I skipped the inherent opinion here, is the argument is backed by exactly nothing.

However, with GPLv3 you can do everything wb wants, there are no good in terms of technical or non-evils reasons to avoid it.

You get 4 freedoms, depending how you invoke or link to it, a LGPL version is a matter of just allowing it, it can carry the exception to allow further LGPL works or not.

In addition to that it is gratis. Putting it, in terms of adoption, into a very lucrative space, that many other formats aren't in.
 
Last edited by a moderator:
From what i understood, the only rational argument I saw on /. likened the problem of increased quality or bandwidth to that of websafe graphics.

Though still reaching a different conclusion to me.

I still remember the discussion behind websafe colours, back then people thought nobody really needed something better, had or would get it, nor the use, and if they did, could similarly deal with the inconvenience. That, along with the question of support, has since been solved.

This assumption hinges on two things, that

  • people who need extra are technical, and that
  • technical users use their skills to ability.
Knowing the skill required isn't for everyone, its a backend problem, which makes it a very top-down question, instead we could even the playing-field.

I would much rather (as a technical user) just put up the source /(best) image I got, and let scaling and quality be left to ability of the users software, without having to make assumptions.

I want to be that user too, on a general basis, the technical part of me cares even more about having the ability. Don't compress quality.

My assumptions on the backend, when I'm just doing something else, will always be limited, or it will become an incomprehensible problem to deal with, with fallbacks and different methods etc, and even then, it still gets outdated.

In total, for however much better the total sum of bandwidth has gotten, we seem to still be able to use it up.

That is a net gain for everyone.

You cant look a gift horse in the mouth if you don't understand the difference between a horse and a trojan.

I've been rewriting the page to try and help wb make the points he put in. There should be something about rendering time too.
 
I'd agree except that people aren't wording it like that, most are claiming it IS dead because it IS GPL3; complete dismissal.  That's the kneejerk I'm talking about, no consideration that it can be relicensed or reimplemented, just a dead fish before it's done.

On another note, I do like the idea of trolling everyone by pronouncing it "fly-f" :p

I've read most of the comments in the linked threads here and a lot of people are saying exactly what I said. Some people are saying it's screwed forever yes, but other people are saying that they want it to be stuck with GPLv3 (much like comradekingu is here)

To be fair, in the real world things that are licensed GPL don't tend to switch later, unless they sell proprietary licenses. The particular circumstances here make this more plausible, and others are pointing that out too. But that's not the most immediately obvious thing in the world.

I can also see why some people won't be that enthusiastic about the recommendation to reimplement under a less restrictive license when right now the technical spec only exists in the GPLed code. You definitely don't want to try reimplementing from that and risk getting called a derivative work.
 
Last edited by a moderator:
A lot of people making the arguments seem to have missed that FLIF still under development and the format isn't stable yet.

-Neelix
 
A lot of people making the arguments seem to have missed that FLIF still under development and the format isn't stable yet.

-Neelix

I guess when you see something publicized like this it's natural to expect it's ready.
 
Back
Top