жёстко. agreed, in general. but all this is true for mainstream development or, opposite, for unique designs. but, if we are making, for example, NDT gauges, with annual production less than 1000 (no one need more!), it does not work. especially, when each customer is known familiar and is not lazy enough to come and complain personally.
2008/1/19, Oleg Verych <[email protected]>: > > @"Sat, 19 Jan 2008 03:41:02 +0300" "Oleg V. Kobrin": > > > do you mind? I do like to match the fenomenon of this community. > > Right, seconded. > > == 2 "GCC for MSP430 - http://mspgcc.sf.net": > > Open Source is all about better technology. This technology is made by > people, who sends patches and maintains code. And only these people have > *rights*. They don't ask for opening closed code, all they ask is > *documentation*. This is the key point. _Code_ and software technology is > business of open source. > > Quality and way open source goes depends on those, who does code. > There can be bad decisions, buggy code, stupid security holes. But, > all this can be solved by others, who will eventually join. > > This doesn't work with Free Software, where word "free" itself is a very > bad choice. FS is about philosophical brainwashing and flame-wars. Many > luzers just like to spend hours and hours by making noise and trolling in > tech forums (like LKML). > > Nearly ten years ago Open Source term was invented to cover long-standing > practice of sharing code for better technology, commercial or free -- > doesn't matter. This is nothing new, only Internet had became more > accessible by many people. > > GNU Compile Collection stands in between of this two. As i mentioned > about Cygnus, GCC wasn't about free, like in Freedom. But RMS is pushing > crap, like ``GNU/Linux'', while GNU is just a very small part of all, > what was done. > > A distro can be perfectly called "x86_64-asus-linux-glibc-perl-debian", > where GCC and GPL are just tools, not goals. > > Technically GCC is going down. Its codebase is gatekeeped by FSF > bureaucrats, and this is why Open BSD people had started bringing back > pcc compiler. > > While language lawyers are flaming about C > > http://article.gmane.org/gmane.linux.kernel/491501 > > some GCC developers accept, that something goes off road > > http://vmakarov.fedorapeople.org/spec/comparison.html > > == > > So, if there isn't documentation about hardware, then one, with open > source hat, must ask about it. > > If TI have no official support of open source software, then maybe they > want to provide best support for it, which is impossible, and, quite > frankly, in true open source, is *not* needed. > > If somebody wants to argue, please, find another place for that crap. > > == My POV about debugging. > > As far as i can see, there's documentation about JTAG and stuff. There is > no info about breakpoints (code steps, etc.). > > While i don't do software development for food, i also don't use > debuggers at all. If i don't have result i want, i try to look hard in > source, to figure out crap i've done. > > Generally, writing tests for every functionality is a good idea. With > microcontrollers this can lead to making additional hardware, to make > sure about many corner cases silicon, connections or metal can give you. > > This requires good knowledge of datasheets, errara, language and tools. > This also requites more time and self organizing, training and order in > the head. And after knowing modern software industry, where fast making > and shipping crap to betatesters-endusers is OK, i think i'd better do > slowly, what i want and need. > > (Debuggers -- it is dealing with consequences, not causes. Hint: Linux > kernel have no debugger in official Linus' tree, while many attempts > to bring kdb were made. Yes some basic functionality is there, but > only because of the pressure.) > > -- > -o--=O`C > #oo'L O > <___=E M > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Mspgcc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/mspgcc-users > > -- _____________ Oleg V. Kobrin
