GIF does support multiple image blocks with positioning and size info. The LZW data is in those blocks. If your LZW data is legal (from a GIF file generated by a legal program) then you can extract the LZW part and build a composite image from other images and you wouldn't be doing any LZW at all. That won't be useful for most image work, but for some things it could be, such as image counter CGIs where each digit comes from a digits file (if your digits are legal).
IMHO, there hasn't been any delay. Unisys has been defending their LZW patent all along. The patent does run out in about 3 years. They have simply changed their course of enforcement to deal with several issues that have been changing. One of those is the rising freeware popularity. I suspect they realize they won't get much out of a non-commercial site using GIFs generated by freeware, but a commercial site is certainly a potential revenue source. And it is legal to collect royalties on the use of a patent, in addition to the implementation of it. And all the while that the use of GIF is growing due to the web growth, Unisys is probably preparing for the coming phase out of the Welch Licensing department. With that, and the lack of future revenue streams, it will soon (before the 3 years are up) not be worth pursuing infringers. So get them for everything for the next 3 years and get it now. And there is the time-value of money gotten sooner, too.
Web counters can easily use PNG or JPEG. Both have libraries available for programmers (for sure in C/C++, surely also in Perl, perhaps even in Python/TCL). You can also generate LZW-free GIF if you want (angif, gd, libungif). Another technique is to extract the LZW data block from an existing GIF generated by a Unisys-licensed program, for each digit, and piece them together into a multi-image-block (angif can help in this part) GIF file. If your digit collection is already in GIF and was produced by legal code (but how would you know?) this way you can build new GIFs without doing any compress or uncompress operations. The LZW patent only covers LZW, and hence only the data portion of the image block of GIF, not the whole GIF file.
One of the disadvantages of chipsets that supported socket 7 at 100 MHz FSB was that they had a caching area size that was small, such as 128meg for the Ali chipset. That puts a ceiling on useful RAM expansion. While most won't need more than that, some do. With the K6-3 doing the cache on it's own now, my question will be what the caching area of this new cache control logic will be. The Intel Celeron in Socket 370, and the Pentium-II L2 cache, have an area confirmed to be 4gig.
Hopefully the K6-3 cache will run at full core speed even with a 100 MHz FSB.
*nix vs those other OSes
on
Love of Unix
·
· Score: 1
Yes, it sucks less.
I've worked with other OSes both before and after Unix. But Unix, and in particular Linux, have simply done the job better, and suited the needs better, than anything else. I do remember when peecees first came out and I read a projection that some day everyone would have a mainframe on their desk. At the time I was working with MVS and thought //DUMB EXEC PGM="NOWAY" would people ever deal with that, not even computer people. VM/CMS was the alternative OS of the day among mainframes.
Well, that day is here, and if we don't have some expensive vendor OS, we have Linux or FreeBSD or Beos or whatever. I love my *nix but I can't say the same about the stupid peecee architecture.
True, you don't need to know C or understand kernel source to admin Linux. But not knowing it does limit the scope of what you can do. It is just like any other skill or understanding. What it does for me (and I've been programming in C since 1982 on over a dozen different platforms) is give me the ability to diagnose problems in programs I am trying to install, and modify those that don't do exactly what I want.
Another example is networking skills. You don't have to understand TCP/IP to admin a system, but it has its advantages if you do know it. And knowing HTML has some advantages, as well:-)
I would suggest learning as many related skills you can, especially if you want to improve your career position even in the sysadmin track.
Slashdot Mirror
GIF does support multiple image blocks with positioning and size info. The LZW data is in those blocks. If your LZW data is legal (from a GIF file generated by a legal program) then you can extract the LZW part and build a composite image from other images and you wouldn't be doing any LZW at all. That won't be useful for most image work, but for some things it could be, such as image counter CGIs where each digit comes from a digits file (if your digits are legal).
IMHO, there hasn't been any delay. Unisys has been defending their LZW patent all along. The patent does run out in about 3 years. They have simply changed their course of enforcement to deal with several issues that have been changing. One of those is the rising freeware popularity. I suspect they realize they won't get much out of a non-commercial site using GIFs generated by freeware, but a commercial site is certainly a potential revenue source. And it is legal to collect royalties on the use of a patent, in addition to the implementation of it. And all the while that the use of GIF is growing due to the web growth, Unisys is probably preparing for the coming phase out of the Welch Licensing department. With that, and the lack of future revenue streams, it will soon (before the 3 years are up) not be worth pursuing infringers. So get them for everything for the next 3 years and get it now. And there is the time-value of money gotten sooner, too.
Web counters can easily use PNG or JPEG. Both have libraries available for programmers (for sure in C/C++, surely also in Perl, perhaps even in Python/TCL). You can also generate LZW-free GIF if you want (angif, gd, libungif). Another technique is to extract the LZW data block from an existing GIF generated by a Unisys-licensed program, for each digit, and piece them together into a multi-image-block (angif can help in this part) GIF file. If your digit collection is already in GIF and was produced by legal code (but how would you know?) this way you can build new GIFs without doing any compress or uncompress operations. The LZW patent only covers LZW, and hence only the data portion of the image block of GIF, not the whole GIF file.
foo!
I just released a C library that generates animated (or if you want, 24-bit true-color that really works on browsers) GIFs that are LZW-free.
The TLD I've seen proposed for nearly 3 years now is .TMK and then you might also do service marks separately with .SMK .
One of the disadvantages of chipsets that supported socket 7 at 100 MHz FSB was that they had a caching area size that was small, such as 128meg for the Ali chipset. That puts a ceiling on useful RAM expansion. While most won't need more than that, some do. With the K6-3 doing the cache on it's own now, my question will be what the caching area of this new cache control logic will be. The Intel Celeron in Socket 370, and the Pentium-II L2 cache, have an area confirmed to be 4gig.
Hopefully the K6-3 cache will run at full core speed even with a 100 MHz FSB.
Yes, it sucks less.
//DUMB EXEC PGM="NOWAY" would people ever deal with that, not even computer people. VM/CMS was the alternative OS of the day among mainframes.
I've worked with other OSes both before and after Unix. But Unix, and in particular Linux, have simply done the job better, and suited the needs better, than anything else. I do remember when peecees first came out and I read a projection that some day everyone would have a mainframe on their desk. At the time I was working with MVS and thought
Well, that day is here, and if we don't have some expensive vendor OS, we have Linux or FreeBSD or Beos or whatever. I love my *nix but I can't say the same about the stupid peecee architecture.
True, you don't need to know C or understand kernel source to admin Linux. But not knowing it does limit the scope of what you can do. It is just like any other skill or understanding. What it does for me (and I've been programming in C since 1982 on over a dozen different platforms) is give me the ability to diagnose problems in programs I am trying to install, and modify those that don't do exactly what I want.
:-)
Another example is networking skills. You don't have to understand TCP/IP to admin a system, but it has its advantages if you do know it. And knowing HTML has some advantages, as well
I would suggest learning as many related skills you can, especially if you want to improve your career position even in the sysadmin track.