CIS/Unysis GIF scam
17 messages in this thread
Personally, I hope this drives people to use something other than GIF, like
JPEG. It's absolutely ridiculous to download 300K 256-color images, when a 60K
JPEG can do it in 24 bits. I don't know how any party could chase down every
developer that's using GIF. I think there are a handful of outrageous claims
in this guy's interpretation of what happened… which simply makes me more
patient to sit back and watch what happens.
It may be coincidental, but I recently heard about a court case where a company
was able to get a decision in their favor based on the premise that their file
format was copyright and protected, and that another party couldn't
reverse-engineer it. In this case, I even think it was in ASCII, not binary.
> Personally, I hope this drives people to use something other than GIF,
I think that's a forgone conclusion. As one of my favourite TV
characters might say it… "The bullet has left the barrel. It is too
late for the shooter to choose another target."
> like JPEG.
JPEG is good for some things, but you wil not find an artist who will agree
that it is any good for his work. While Mom or Pop might snapshot with the
Instamatic, the pro with the Hasselblad sure wouldn't use one.
-larry
If we were to sort all the GIF files in existence into big bins
according to category, just how big do you think that "sex" bin would
be? Or regarding your question, how big would the "mom and pop" bin
be?
Have you read the actual text of the developer's agreement from
Compuserve? Unless I'm the victim of a cruel joke, it was posted to
comp.graphics in the past day or two, and there's a *bizarre* section
1.5 that says that the developer has to insure that the users and
customers of their GIF-based product must use CompuServe as their
information service.
Here's that section…
1.5. Developer shall not grant any customer the right to use a
Product until such customer has been registered by Developer as a
user of the Product and customer's rights to use such Product are
governed by an agreement with Developer providing that (a) the
customer's use of such Product will be primarily for accessing the
CompuServe Information Service and for manipulating and viewing data
received through the CompuServe Information Service, and (b) the
customer will not alter, enhance or redistribute any Product.
Perhaps someone is quoting some *other* type of agreement that Compuserve has
with developers who make products that work with CIS, as opposed to other
systems.
Also, re: your argument about JPEG versus GIF, I don't see many Hasselblad
users expecting great results from 256-color images. Or do they?
> … JPEG versus GIF, I don't see many Hasselblad users expecting
> users expecting great results from 256-color images. Or do they?
Certainly not, but then the Hasselblad vs. snapshot was an analogy only, likely
not a great one, since it looked like I was talking about those who look to a
graphics format to save their photos.
No, I was referring to the artist, who creates a work using a program, and who
saves it in a format suitable for what he is doing. GIF is fine for this
purpose, and the artist will see no problem with a 256 colour limitation,
because he has chosen to work with that limitation, or he would not have saved
it that way. When the intended audience views the picture, he sees what the
artist created. With JPEG, he sees what the artist created, filtered through a
lossy file format, and the end result is an approximation.
-larry
G'day,
PMFJI,
>> With JPEG, he sees what the artist created, filtered through a lossy
>> file format, and the end result is an approximation.
JPEG does not equal "Lossy". JPEG has many modes, the most familiar to us is
the "90% reduction in file size – lossy mode" but there are non-lossy modes in
the JPEG specification.
Sam Yates Montrose, Victoria,Australia.
samyates@werple.mira.net.au
> but there are non-lossy modes in the JPEG specification.
Are there? I have not seen any mention of "non-lossy" JPEG modes; only of
degrees of loss.
-larry
G'day,
>> Are there? I have not seen any mention of "non-lossy" JPEG modes; only
>> of degrees of loss.
I have been informed by reliable sources that there is at least one
"Professional JPEG" mode that is lossless. The resulting file is no where near
as small as the lossy modes, but all the data is there.
I had a list of all the JPEG modes somewhere, I'll see if I can dig it out. (It
may have gone south in the last hard drive purge).
As far as I know there is no "non-professional" software that
supports the non-lossy mode(s) currently availible.
Sam Yates Montrose, Victoria,Australia.
samyates@werple.mira.net.au
> I have been informed by reliable sources that there is at least
> one "Professional JPEG" mode that is lossless.
That's what I thought, too, but have yet to see it. A lot of other folks are
looking for it, too, so if it's out there, it's well hidden.
-larry
>>With JPEG, he sees what the artist created, filtered
>>through a lossy file format, and the end result is an approximation.
Actually JPEG is A LOT better than GIF for the artist. Since JPEG quality is
creator selectable, therefore the artist can decide what he wants. If he want
to have people see their creation without giving the original away, they can
use a lossy compression otherwise they can use JPEG non-lossy options. So is
closer to non-digitsl art, the difference of giving someone the original or
photocopy (or photograph, or reproduction) of the original.
Eduard Joseph
Well, it's my understanding that there are only varying degrees of accuracy in
recovery, and that so far, there are no lossless formats for JPEG.
-larry
Sorry to ask but where are you getting this. YES, there is a lossless
compression in the JPEG format it is descrived in the JPEG specs it is coded in
the JPEGproc = 14 (consistant with the ISO DIS 10918-2). The lossy compression
very popular out there uses JPEGproc = 1.
The lossless compression uses a special prediction algoritm.
The one thing it can not have is DCT algorithms since they add error (making
it always lossy).
This has been in available at least since late 1992. And is no secret. It is
also explained in the TIFF 6.0 specs (which is used for its JTIFF) which have
been available since June, 1992.
Where have you been looking they are not HIDDEN at all it has it's own section,
JPEG code, you name it. 🙂
Eduard Joseph
I'll take a look for it. Thanks for the pointers.
-larry
Well, I grabbed a JPEG spec, and it says absolutely nothing about any lossless
mode.
I also grabbed a TIFF spec, and am printing it out as I type. If the mention is
in this, then it is no wonder I didn't see it.
-larry
Where did you get the JPEG specs from C-Cube?
Eduard
I agree John, GIF is a lousy format for displaying images, especially in
this day and age of 24-bit graphics boards, AGA, etc. The guys
interpretation might be a little skewed, but I read the developer's
agreement myself and it is very vague in it's specifications for
"disposition" and "field of use".
My feeling is that Unisys and CompuServe don't have much of a leg to stand
on. GIF has been use in the public domain for years and to suddenly
indicate that everyone will now have to pay royalties, well I wish them
luck in court.
// John (Warp 4040 & Emplant Quadra 840)
\X/ Mercury@ins.infonet.net