For development, I write C code which is converted by a tool so that it compiles and work under Windows as a binary plugin for a simulator. The production system is an OS-less Big-Endian microcontroller.
Since the simulator is running on a little-endian PC, the tool we wrote has to modify all the code so that it stores its working variables in Big-Endian format and converts it before use. It also has some cleverness to convert all the bit-fields so that their storage and access is the same on both systems. The code has to do a lot of IO with other systems (both simulated and not) so internally storing the data in Big Endian format on the Little Endian system prevents us from having to change the code much or adding special purpose Endian converters on all the IO interfaces. Since the production system is a few orders of magnitude slower than the simulator we don't care about wasting extra cycles there. Using this approach, we were able to convert the Big Endian code to work under simulator pretty quickly.
Perhaps the extra formats are considered arcane and unuseful, but the extra compression with regular zips, and even more so with the 7zip format, is worthwhile.
The newest eterm's have experimental screen support. It allows you to use something like tabs, where each one is a new window in a single screen session making it by far the coolest terminal. Experimental means it has a tendency to crash occasionally, but you don't lose your work because screen is still running, so you just have to restart eterm.
The problem here is that a fundemental practice of open-source is to allow anonymous access to development sources. You can't provide anonymous cvs or subversion access through ssh.
One of the issues here that this "modularized" distribution is trying to account for is that you want a machine that has what you need and only what you need. Do you need a nameserver? Fine install a base linux system and bind. Do you need a nameserver with a database backend? Now you need bind + (mysql | postgres). So now you have the question does your bind package come compiled with postgres support or not? Either way, you've violated the spirit of this modularity.
Source distributions can have "flavors (BSD)" or "use flags (Gentoo)". I'm not arguing that compiling everything from scratch is most practical, but Gentoo already fulfills, and probably exceeds, the flexibility goals of the article.
All it would need to be kick-ass base for this project would be a 1) a decent gui for installation and updating, 2) a binary server that required you to re-compile only when you wanted something fairly unique.
I just don't understand how this is a bad thing. One of top guys in my company just plain doesn't trust open source and never will. The SCO vs. IBM thing has given fuel to the fire of his arguments. Things like this help out out an open-source project and everyone using it.
First of all there's the direct benefit to those who are indemnified. Then there are all the secondary benefits:
Everyone knows that a company is ready to stand up for this product, reducing it's "easy target" status.
Everyone knows that this company has enough faith in the IP of this product to put their money where their mouth is. Self-assurance is important to those who judge by appearance.
JBoss group should get more money from those who need this peace of mind increasing the amount of money they have to invest in making JBoss better.
A wider user base increases the chance that other organizations (EFF etc.) will help out in a heated legal battle.
And the great part is that all the secondary benefits apply towards all those of us who don't really feel like paying for what other people worked hard to make. I say we cheer JBoss on for that move.
Not necessarily.
Everybody assumes that assembly makes the fastest programs. The reality is that optimizing C compilers can produce faster code than most programmers. Even among those who have knowledge to write faster assembly, the sheer time involved can make it unworthwhile to do something optimally. After writing something at a snail's pace for days, the developer often says to himself: I'll come back later and code this up as a binary tree, for now I'm just going to use a linear search and get it done.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 1
Our patent system exists to encourage our capitalistic system and to make it work better. One example where it works against this is in the ability to use a patent, or copyright laws, to lock up any kind of interface. Should printer companies give away their printers and make money gouging people for ink since no-one else can sell ink for their machines? Some might not see the problem with that; but in the end prices are less fair for consumers and higher in the long run.
Software patents aren't necessary whenever code is distributed in binary form, because being a patent it should be novel and show some actual work in building on previous concepts. The only real use for them that I have seen is to control interface access to increase monopoly power beyond an entity's creative efforts.
You could write a proxy (and optionally make it transparent) of your own that replaced the page name in HTTP requests with the text version of the IP address you were sending to. It's not a very elegant solution; going through two proxies to achieve the effect of going through none, but at least it would work. Alternatively, you could do what I would do in your situation, piss and moan to my ISP until they fixed it.
Slashdot Mirror
For development, I write C code which is converted by a tool so that it compiles and work under Windows as a binary plugin for a simulator. The production system is an OS-less Big-Endian microcontroller. Since the simulator is running on a little-endian PC, the tool we wrote has to modify all the code so that it stores its working variables in Big-Endian format and converts it before use. It also has some cleverness to convert all the bit-fields so that their storage and access is the same on both systems. The code has to do a lot of IO with other systems (both simulated and not) so internally storing the data in Big Endian format on the Little Endian system prevents us from having to change the code much or adding special purpose Endian converters on all the IO interfaces. Since the production system is a few orders of magnitude slower than the simulator we don't care about wasting extra cycles there. Using this approach, we were able to convert the Big Endian code to work under simulator pretty quickly.
Perhaps the extra formats are considered arcane and unuseful, but the extra compression with regular zips, and even more so with the 7zip format, is worthwhile.
The newest eterm's have experimental screen support. It allows you to use something like tabs, where each one is a new window in a single screen session making it by far the coolest terminal. Experimental means it has a tendency to crash occasionally, but you don't lose your work because screen is still running, so you just have to restart eterm.
I think that goes in the category of "Sad but true".
The problem here is that a fundemental practice of open-source is to allow anonymous access to development sources. You can't provide anonymous cvs or subversion access through ssh.
One of the issues here that this "modularized" distribution is trying to account for is that you want a machine that has what you need and only what you need. Do you need a nameserver? Fine install a base linux system and bind. Do you need a nameserver with a database backend? Now you need bind + (mysql | postgres). So now you have the question does your bind package come compiled with postgres support or not? Either way, you've violated the spirit of this modularity. Source distributions can have "flavors (BSD)" or "use flags (Gentoo)". I'm not arguing that compiling everything from scratch is most practical, but Gentoo already fulfills, and probably exceeds, the flexibility goals of the article. All it would need to be kick-ass base for this project would be a 1) a decent gui for installation and updating, 2) a binary server that required you to re-compile only when you wanted something fairly unique.
- Everyone knows that a company is ready to stand up for this product, reducing it's "easy target" status.
- Everyone knows that this company has enough faith in the IP of this product to put their money where their mouth is. Self-assurance is important to those who judge by appearance.
- JBoss group should get more money from those who need this peace of mind increasing the amount of money they have to invest in making JBoss better.
- A wider user base increases the chance that other organizations (EFF etc.) will help out in a heated legal battle.
And the great part is that all the secondary benefits apply towards all those of us who don't really feel like paying for what other people worked hard to make. I say we cheer JBoss on for that move.Not necessarily. Everybody assumes that assembly makes the fastest programs. The reality is that optimizing C compilers can produce faster code than most programmers. Even among those who have knowledge to write faster assembly, the sheer time involved can make it unworthwhile to do something optimally. After writing something at a snail's pace for days, the developer often says to himself: I'll come back later and code this up as a binary tree, for now I'm just going to use a linear search and get it done.
Our patent system exists to encourage our capitalistic system and to make it work better. One example where it works against this is in the ability to use a patent, or copyright laws, to lock up any kind of interface. Should printer companies give away their printers and make money gouging people for ink since no-one else can sell ink for their machines? Some might not see the problem with that; but in the end prices are less fair for consumers and higher in the long run. Software patents aren't necessary whenever code is distributed in binary form, because being a patent it should be novel and show some actual work in building on previous concepts. The only real use for them that I have seen is to control interface access to increase monopoly power beyond an entity's creative efforts.
You could write a proxy (and optionally make it transparent) of your own that replaced the page name in HTTP requests with the text version of the IP address you were sending to. It's not a very elegant solution; going through two proxies to achieve the effect of going through none, but at least it would work. Alternatively, you could do what I would do in your situation, piss and moan to my ISP until they fixed it.