Multi-Platform Video Codec Seeks New Home

Eric Smith asks: "I own a next-generation video codec development company, idea65 (covered on Slashdot as our previous incarnations Opencodex, and Flashingyellow). We have a finished product, but don't quite know what to do with it, and we're looking for someone (or some company) interested in taking it over." Here's the chance for some of you out there looking for a good cause, to contribute something. Loads of people (me, included) would love a good cross-platform video codec.

We started our journey as an open-source project contest in response to DivX, before DivX networks came into being. Due to a variety of issues (not the least of which was our main investor pulling out and funding having to come out of my own pocket), we mutated into a closed-source project that we intended to distribute ourselves through the help of a third party. We finished product development almost a year ago and have a really great portable video codec that runs on Linux, Mac OS, and Windows.

The problem we've run into is that with the economy being as it is, our candidates for distribution assistance have also all dried up. We've considered just GPL'ing it and seeing what the open-source community could do with it, but don't have anyone to oversee changes and official versions, not to mention from the looks of the DivX 4.0 project, there don't seem to be a lot of people interested in (or with the knowledge to) work on video codecs.

More or less, we've got a bunch of very well written CodeWarrior projects that need to find a new home as we don't really have the expertise or financing to sell it or even give it away. So, I'm interested in knowing if anyone has any suggestions for what to do with the project, or interest in taking it over (those with experience with this kind of thing)."

If seriously interested, you can contact Eric using the mailto link at the beginning of this article.

Comments

Ogg Tarkin

[Shrubbman]

If you're codec is patent free, the people over at the ogg multimedia project would be bigtime interested. They've got the audio portion (vorbis) well along but they're still aways away from having their video portion (tarkin) completed. Head over here for more info

The problem is...

[easter1916]

We have a finished product, but don't quite know what to do with it

Prior to developing the product and forming a company, did anyone mention the words "business plan" to you?

So basically...

[glowingspleen]

So basically you just used Ask Slashdot to market a product or further your career.

Just kidding, everyone does that.

quality?

[crayz]

Would it be possible to show us the quality of this codec somehow? Maybe start with uncompressed video, and have the DivX, VP3, etc. people all compress it down to a specific bitrate, and you do the same. Then once it's compressed down to that bitrate you could decompress it(since we don't have a decoder...), thus letting the general public see the quality of your codec.

Would that work? Because there are a lot of codecs, and unless you can show that this one is better than the others, I really don't see why people would be interested.

It all depends...

[Junta]

You mention that now that DivX 4.0 is out, no one cares anymore. Does your codec outperform DivX4.0 in terms of bitrate, quality, and/or performance? If the answer to all the questions is no, then perhaps it would be best to let it die. If the answer is "not now, but with work it could be..." and you want to stop working, you may want to first fish around for interested companies, and as a last resort give it to the Ogg group as something they could hammer into being Tarkin. I guess if you really don't care about the code anymore but really don't want to let it die, you could pass it on to the Ogg people anyway and they can decide for themselves whether the code is worthwhile or at least salvagable. If it isn't, then they can kill it instead of you :)

Re:Don't be so quick to GPL!

[mr3038]

Don't GPL it, that would be silly. If you guys made the darn thing with the intention of earning money, you should darn well get some money for it!

Well, if you intent to get some money with it GPLing shouldn't ruin the plan. If somebody wants to use this codec commercially they probably sell closed source program with it and cannot therefore use codec without purchasing different license. GPL doesn't restrict from releasing product under another product simultaneously. One could even claim that GPL version would be a full-featured demo to sell codec.

If the codec does something revolutional like not using DCT and interpolation between keyframes then not to GPL it may be a good idea because ideas aren't restricted by copyright. Without money you cannot patent it and that would be only yet another hated software patent anyway.

Whether or not the use of GPLed codec would be legal in Windows or MacOS is another question. Most programs in these platforms are closed and cannot therefore link with GPLed code. Strictly interpreted this means that you cannot use GPLed codec in say for example WMP. On the other hand WMP may be claimed to be part of OS and GPL allows linking with OS libraries...

An idea...

[Fnkmaster]

Since really the only way as I understand it that such a product could make money is embedding in successful closed source projects, I think this would be an ideal candidate for the TrollTech business model. Release a GPL version that is embeddable only into GPL and GPL compatible Free Software products. This will gain you recognition and acclaim if your product is good. Now you've stirred up interest. You should release, simultaneous with the GPL release, a press release indicating your intention to issue commercial licenses for embedding in closed source products. Hell, you can even claim your commercial licensees will get access to the "Plus" form of your codec.

Now pimp the hell out of the GPL version and everyone that whines about how they can't use it in their closed source products, point them to your web page explaining how to contact you/your company for commercial licensing terms. Also perhaps consider a joint distribution agreement up front with some commercial video tool providers whereby you will develop plug ins, etc. I gather your point is this might be hard in this climate. Also consider getting pr0n distributors to use it. No, I'm not kidding. Honestly, people download pr0n from usenet, etc. If there is pr0n out there in your video format, people will get players for it. This will eventually convince commercial users that your format is worthwhile. There are plenty of Linux/Mac pr0n viewing folks out there, so you definitely will find some rapid fans if you take this tactic.

Anyway, this is my advice for a bootstrapped marketing technique that you might find effective. I make no promises, but it sounds like you don't have much to lose if you are posting to Ask Slashdot for marketing advice (hint: lots of /. readers can tell you about GPL violations, some can tell you about how the Linux kernel is put together, and very few can tell you how to successfully market a product).

Re:Wow - what a business plan

[misterye]

Actually, we did have a very detailed business plan, that we thought was actually going to make money and contribute something to the open-source community. But, as with many things, and as someone said earlier, don't depend on third parties. We didn't have much choice but to rely on third parties for distribution and funding. In the end, all the funding came out of my pocket, and the distributor went under. We have a good product, we actually beat DivX to being done by about 3 months, but we got stuck there. The codec is very light and highly portable, the original version was less than 16k compiled and out-performed DivX and was patent and copyright free.

Though I guess that is something that people will want to see for themselves, and that's where we are stuck, we don't even have the funding for the legal help to get it into testers hands. After working on this as hard as I have, I would hate to see the project just die w/out it seeing the light of day. That's why I asked for possible answers from Slashdot. Maybe someone can think of something we haven't. Its happened before.

The only think that makes your CODEC worthwhile...

[oddityfds]

Possibly, the only thing that makes your CODEC worthwhile is if it is patent free. The only CODEC to date that I know of that is free of software patents is the H.261.

If we're going to use an algorithm encumbered by patents, we might as well use MPEG-4.

However if your CODEC is not covered by any patents, then please consider releasing it under a BSD or GPL license.

For information on why software patents are bad for free software, please visit The League for Programming Freedom

I'm sure we can fit you in somewhere...

[mcelrath]

I'm sure there's room around here...maybe under the couch? Maybe between 3ivx and DivX? Or between MPEG-4 and Sorensen? If not then surely there's room between MJPEG and Indeo? Oh, I feel like breaking into song over the wonderful video codec situation!

Oh! 3iv1 3iv2, aasc abyr and aemi too! afli, aflc boy those are old! AMPG, ANIM, AP41 and you! Think of how your home videos will look, in ASV1, ASV2 or ASVX!! Mine look great, and I'm sure yours will too, with AUR2 or AURA!

And out of the A's and into the B's la da da do do do deeee! Bink, bt20, btcv bw10, boy those b's are short! Those B codec makers better get a move on! But onto the c's like the birds and the bees, cc12, cdvc, cfcc, cgdi, cham, cjpg, cpla, cram, cvid cwlt, cyuv, cyuy! Boy these things multiply fast! Makes me wonder, why the're called video codecs, and not WABBITS!

But D's come along, DIV2 and DIV4 and DIV5, with venerable DIVX short behind. DMB1, DMB2, DSVD, DUCK, DVAN, DVSD, DVE2, DVX1, DVX2, DVX3, DXTN, DXTC, and no more D's do we have! And the E's are short, because normal people don't start video codecs with vowels, ETV1, ETV2, ETVC are all that survive.

Only 3 F's, because F stands for Flunk, FLJP, FRWA, FRWD are fun! Oh my! GLZW, GPEG, GWLT from Microsoft? But videoconferencing still lives H260 goes plop, followed by H261, and H262, H263, H264, H265, H266, H267, H268, But finally everyone knows H269! HFYU, HMCR, and HMRR round out the H's!

(Egad, am I done yet!)

Not hardly buddy! There's I263, IAN, ICLB, IGOR, IJPG, ILVC, ILVR, IPDV, IR21, IV30, IV31, IV32, IV33, IV34, IV35, IV36, IV37, IV38, IV39, IV40, IV41, IV42, IV43, IV44, IV45, IV46, IV47, IV48, IV49, and IV50 rounds out Intel's evil contribution! But wait! There's more! Call now and you'll receive this free JBYR, JPGL, KMVC, and LEAD, LJPG.

Not to worry, M is here! Here are all the ways Micro$oft can fuck a standard! M263, M261, MP42, MP43, MP4S, MPG4, MRLE, MSVC Oh my! I like MJPG, cause my marvel uses it, but Matrox also has MTX1, MTX2, MTX3, MTX4, MTX5, MTX6, MTX7, MTX8, MTX9! More M's! mJPG is not the same as MJPG? MCAM, MC12, MPEG, MRCA, MWV1, nAVI, NTN1, NVS0, NVS1, NVS2, NVS3, NVS4, NVS5, NVT0, NVT1, NVT2, NVT3, NTT4, NVT5, PDVC, PGVV, PIM1, PIM2, PIMJ, PVEZ, PVMM, PVW2, qpeg, QPEG, RGBT, RLE, RT21, rv20, rv30, RVX, s422, SDCC, SFMC, SMSC, SMSD, smsv, SPIG, SQZ2, SV10, STVA, STVB, STVC...oh god I'm getting bored...STVX, STVY, SVQ1, TLMS, TLST, TM20, TM2X, TMIC, TMOT, TR20 TSCC, TV10, TY2C, TY2N, TY0N, UCOD, ULTI, V261, VCR1, VCR2, VDOM, VDOW, VDTZ, VGPX, VIFP, VIDS, VIVO, VIXL, VLV1, VP30, VP31, VX1k, VX2K, VXSP, WBVC, WHAM, WINX, WJPG, WNV1, x263, XLV0, XMPG, XXAN, Y41P....almost there! Y8, YC12, YUV8, YUV2, YUYV, ZLIB, ZPEG!

And that ends my really bad song. But wait! Thre's more! Those are only the ones with FOURCC definitions! That doesn't include file types! There's MPEG-PS, AVI, Quicktime, and the venerable Microsoft format heist asf.

I think the statement "I developed a new video codec!" should be punishable by death.

--Bob