MS has every right to do this with BSD licensed code. And they do with GPLed code to, but if it was GPLed code then they would have to release the source to the derivative product under the GPL.
Note that I'm not making any statements for or against either license, or for or against MS. I'm just pointing the key the difference in these popular licenses.
> Word 97 is a perfectly adequate word processor. So was Word 95 for that matter.
I have to disagree here. My experience with Word 95 was that it used to frequently write out.doc files that it could not read. This seemed to happen often enough that I ended up saving to a new filename every single time I saved. (Nice version control, huh?)
Word 97 was a dramatic improvement in this regard, but even still I see this occasionally with Word 97. The interesting thing is that sometimes once you've generated one of this "sick".doc files, Word 2000 can read it in, and occasionally OpenOffice can too.
Admittedly my experience with Word 2000 is limited (and with XP nil), but Word 2000 appears to me to be better about reading suspect.doc files than 97 - hopefully it's better about writing valid files too.
For documents for internal use only, I generally now will use latex or OpenOffice depending on what the document is for. Unfortunately, the rest of the world wants the ubiquitous.doc format and OpenOffice 1.0 does not cut it for generating.doc files that can be read in and displayed correctly by the different versions of Word; hence the fallback is to use the "real thing" - albeit on Crossover Office.
One book I like is by Janet Ruhl, and the info was gleaned from Usenet back "in the day". It's called The Computer Consultants Guide. I found it to have lots of good information on a variety of subjects, although it is a bit dated.
As I mentioned in another reply, currently one of my domains is suffering from the same problem. Your question inspired me to mull over what one could do a little more. I was thinking since we host our own DNS, we could put in ACLs in our bind setup to disallow queries to the affected domain from the netblock that the spammer is operating from, and perhaps the first level of smtp servers that they are using. (If those are consistent.) This might provide a way to selectively DOS the people who are generating the spam...
We have had the same issue, unfortunately. I asked on the debian-isp mailing list about it and the only real suggestion was to report the spammer in question to their ISP, which I believe to be in Russia.
The long and short of it is that we couldn't do much about it, other than try to minimize the resource waste. In our exim configuration we turned on "receiver_verify" in our exim configuration, which means before the incoming message enters the delivery phase, it's verified that there is a valid receiver. (Before doing this, the incoming message would run through spamassassin and then generate a bounce, using CPU time, memory, etc.) I know it's not much; I hope someone comes up with more suggestions.
> exim tends to choke pretty seriously on some of my mail, though... (fetchmail'd from an Exchange server)
The FAQ talks about exim/fetchmail, if you're seeing the "standard problem" it's not hard to deal with. Probably it would be better to ask on a mailing list if it's not in the FAQ.
Ask on Debian-consultants for information on companies that will contractually agree to provide phone support. If you think about it, it's where the whole open source "thing" is going to end up. Ultimately, RedHat doesn't own the code any more than a Debian developer. (Both RH and the developer DO contribute to the codebase, of couse, RH is a good player with respect to the OS community IMO.)
If you want to be overly simplistic, RedHat is just another company willing to support Linux.
> You said, "Already Mosquitos and flies started > showing up in various places where they were > never seen before."
> Really. I'd like to know what place that is.
What I've seen about it talks about how they are moving to higher elevations on mountain ranges than they have been previously. Here you go.
The other thing I've recently been reading about (not directly related to a warming climate) is that species of mosquitoes have been migrating due to global trade, riding in containers or whatever.
The semantics of concurrency are well defined for VHDL simulation (as long as you're sane and avoid global shared variables, although those are fixed in the 200x LRM as well.) Of course, even with well-defined semantics you can implement systems with deadlocks and race conditions, and VHDL doesn't have high-level primitives like monitors to help you deal sanely with some of the issues.
And you're right that for synthesis there are whole rats-nests of tool-specific issues. The state of HDL synthesis is quite disappointing and seems to have ground to a halt.
I have no experience with Verilog but what I have heard agrees with your assertions very well. I'll check out Handel-C sometime.
It sounds like an encrypted filesystem would be a better way to go to me. Then you wouldn't have to worry about hiding the dongle where the jackbooted thugs can't find it.
> Do you know of any language or system that got this right?
Well, VHDL, the hardware description language, has a formal semantics defined (in a book written by some friends of mine.) It also addresses concurrently explicitly in the standard, which is nice.
It's far from a useful general purpose programming language though.
> You've run MS-Office on Linux? It was faster? Have you tried it with one of the newer 2.5.x kernels, you may notice a speed increase.
I run MS Office on Linux all of the time - see Codeweavers for info. BTW, Office on CXOffice is definitely faster than OpenOffice. But I tend to use OpenOffice for everything that I can, given that it's free and all, and world domination would be a Good Thing in my opinion.
Often when several of my colleagues and I are working on a paper together, we will use LaTeX + CVS. It works very well, the merging and conflict resolution work well with latex. A couple of important things to make it smoother: 1) Make sure everyone has their editors set to the same word wrap. This is very very very important so you don't get artificial conflicts. 2) You can split your tex across multiple files if you want to make the chances of conflicts less likely. 3) If you want good PDF output in the long run, read about pdflatex and make sure you write tex that it can deal with. Pdflatex generates pdf that is searchable, hyperlinked, etc, unlike dvipdf. It is far superior to dvipdf in every way, and worth the trouble of learning about.
If you use latex anyways, this is a great way to collaborate. If you're working with people who would rather use Word, well, then this isn't too helpful;-)
One last alternative is to write text files, control them with CVS, and then when the writing is done, pull them into Word for formatting. I have worked with people this way too. It's a pain with respect to figures and all of that, but it's a good way to ensure consistent styles, reference and footnote numbering, etc.
I've had several over the years, and they are excellent. Even better, IBM semi-officially supports Linux on the models that I've owned... (I.e. there are FAQs on IBM's support site discussing how to run Linux on the models I've used.)
My company has a similar setup, but we support sending mail. A typical session looks like this: >~ telnet aol.com smtp Trying 192.168.10.1... Connected to mx1.aol.com. Escape character is '^]'. 220 mx1.aol.com ESMTP Exim 3.35 #1 Thu, 03 Jul 2003 20:27:34 -0400 helo cliftonlabs.com MAIL FROM: dmartin@foobar.fake RCPT TO: abuse@aol.com (and so on.)
Sadly, I'm probably the only slashdotter old enough to remember when you could forge email that way;-)
> I put up with the clunkiness until KDE 3.1 is stable enough for testing.
Packages for woody are available from download.kde.org. While they aren't nearly as good as official.debs (they don't upgrade easily, for instance), they work extremely well and I have had no stability issues on them on desktop machines that use woody.
The other thing that I should have pointed out by now is that there are various backport archives for many common applications, where you can have a stable "core" but run the latest and greatest applications easily. I happen to use Adrian Bunk's backports but there are others too. The/etc/apt/sources.list line for Adrian Bunk's backport collection is: deb http://www.fs.tum.de/~bunk/debian woody/bunk-1 main contrib non-free
YMMV, but I have had very few problems using these backports, and I have used them on many machines.
> So, given Debian's release cycle, that's like, what? One upgrade? Over the past 7 years?
I assume you're being sarcastic, but... There are lots of security updates, so upgrading for those is a fairly regular issue. (The same issues affect RH too, of course.) And given that there's a major release every 18-24 months currently, more frequently in the past, there have been many major upgrades. Note that the machine in question is a server, so not having frequent major upgrades suits me just fine.
On the other hand, my main development box and desktop machine runs the unstable archive, and is usually on the bleeding edge. (Currently it has KDE 3.1.x and mozilla-firebird, g++ 3.3, etc. running on it.)
Slashdot Mirror
MS has every right to do this with BSD licensed code. And they do with GPLed code to, but if it was GPLed code then they would have to release the source to the derivative product under the GPL.
Note that I'm not making any statements for or against either license, or for or against MS. I'm just pointing the key the difference in these popular licenses.
Looks like he goes to the finest unaccredited University out there.
(Laugh, it's a joke -m aybe they are accredited, I have no clue!)
> Word 97 is a perfectly adequate word processor. So was Word 95 for that matter.
.doc files that it could not read. This seemed to happen often enough that I ended up saving to a new filename every single time I saved. (Nice version control, huh?)
.doc files, Word 2000 can read it in, and occasionally OpenOffice can too.
.doc files than 97 - hopefully it's better about writing valid files too.
.doc format and OpenOffice 1.0 does not cut it for generating .doc files that can be read in and displayed correctly by the different versions of Word; hence the fallback is to use the "real thing" - albeit on Crossover Office.
I have to disagree here. My experience with Word 95 was that it used to frequently write out
Word 97 was a dramatic improvement in this regard, but even still I see this occasionally with Word 97. The interesting thing is that sometimes once you've generated one of this "sick"
Admittedly my experience with Word 2000 is limited (and with XP nil), but Word 2000 appears to me to be better about reading suspect
For documents for internal use only, I generally now will use latex or OpenOffice depending on what the document is for. Unfortunately, the rest of the world wants the ubiquitous
One book I like is by Janet Ruhl, and the info was gleaned from Usenet back "in the day". It's called The Computer Consultants Guide. I found it to have lots of good information on a variety of subjects, although it is a bit dated.
As I mentioned in another reply, currently one of my domains is suffering from the same problem. Your question inspired me to mull over what one could do a little more.
I was thinking since we host our own DNS, we could put in ACLs in our bind setup to disallow queries to the affected domain from the netblock that the spammer is operating from, and perhaps the first level of smtp servers that they are using. (If those are consistent.) This might provide a way to selectively DOS the people who are generating the spam...
We have had the same issue, unfortunately. I asked on the debian-isp mailing list about it and the only real suggestion was to report the spammer in question to their ISP, which I believe to be in Russia.
The long and short of it is that we couldn't do much about it, other than try to minimize the resource waste. In our exim configuration we turned on "receiver_verify" in our exim configuration, which means before the incoming message enters the delivery phase, it's verified that there is a valid receiver. (Before doing this, the incoming message would run through spamassassin and then generate a bounce, using CPU time, memory, etc.) I know it's not much; I hope someone comes up with more suggestions.
> exim tends to choke pretty seriously on some of my mail, though... (fetchmail'd from an Exchange server)
The FAQ talks about exim/fetchmail, if you're seeing the "standard problem" it's not hard to deal with. Probably it would be better to ask on a mailing list if it's not in the FAQ.
> Debian is the LAST thing these people need.
Are you going to back this up with some facts, or are you touting your own religion?
Ask on Debian-consultants for information on companies that will contractually agree to provide phone support. If you think about it, it's where the whole open source "thing" is going to end up. Ultimately, RedHat doesn't own the code any more than a Debian developer. (Both RH and the developer DO contribute to the codebase, of couse, RH is a good player with respect to the OS community IMO.)
If you want to be overly simplistic, RedHat is just another company willing to support Linux.
> You said, "Already Mosquitos and flies started
> showing up in various places where they were
> never seen before."
> Really. I'd like to know what place that is.
What I've seen about it talks about how they are moving to higher elevations on mountain ranges than they have been previously.
Here you go.
The other thing I've recently been reading about (not directly related to a warming climate) is that species of mosquitoes have been migrating due to global trade, riding in containers or whatever.
Should "everyone" have access to the derived works? Apparently you don't think so, but many disagree.
The semantics of concurrency are well defined for VHDL simulation (as long as you're sane and avoid global shared variables, although those are fixed in the 200x LRM as well.) Of course, even with well-defined semantics you can implement systems with deadlocks and race conditions, and VHDL doesn't have high-level primitives like monitors to help you deal sanely with some of the issues.
And you're right that for synthesis there are whole rats-nests of tool-specific issues. The state of HDL synthesis is quite disappointing and seems to have ground to a halt.
I have no experience with Verilog but what I have heard agrees with your assertions very well. I'll check out Handel-C sometime.
It sounds like an encrypted filesystem would be a better way to go to me. Then you wouldn't have to worry about hiding the dongle where the jackbooted thugs can't find it.
> Do you know of any language or system that got this right?
Well, VHDL, the hardware description language, has a formal semantics defined (in a book written by some friends of mine.) It also addresses concurrently explicitly in the standard, which is nice.
It's far from a useful general purpose programming language though.
Follow the link, the parent IS insightful.
1 mouse 1 screen 10 pcs is what you'll get if you set it up right.
I've had no problems with it so far. I believe I've had it for a year.
> You've run MS-Office on Linux? It was faster? Have you tried it with one of the newer 2.5.x kernels, you may notice a speed increase.
I run MS Office on Linux all of the time - see Codeweavers for info. BTW, Office on CXOffice is definitely faster than OpenOffice. But I tend to use OpenOffice for everything that I can, given that it's free and all, and world domination would be a Good Thing in my opinion.
Often when several of my colleagues and I are working on a paper together, we will use LaTeX + CVS. It works very well, the merging and conflict resolution work well with latex. A couple of important things to make it smoother:
;-)
1) Make sure everyone has their editors set to the same word wrap. This is very very very important so you don't get artificial conflicts.
2) You can split your tex across multiple files if you want to make the chances of conflicts less likely.
3) If you want good PDF output in the long run, read about pdflatex and make sure you write tex that it can deal with. Pdflatex generates pdf that is searchable, hyperlinked, etc, unlike dvipdf. It is far superior to dvipdf in every way, and worth the trouble of learning about.
If you use latex anyways, this is a great way to collaborate. If you're working with people who would rather use Word, well, then this isn't too helpful
One last alternative is to write text files, control them with CVS, and then when the writing is done, pull them into Word for formatting. I have worked with people this way too. It's a pain with respect to figures and all of that, but it's a good way to ensure consistent styles, reference and footnote numbering, etc.
I've had several over the years, and they are excellent. Even better, IBM semi-officially supports Linux on the models that I've owned...
(I.e. there are FAQs on IBM's support site discussing how to run Linux on the models I've used.)
I think Everytown, USA, would be filled with gawdy advertising of various forms.
Hopefully there would be some useful innovations, too, but most likely it would promote insane inefficiency. (Think Ford Expedition*10...)
"It's easier to get forgiveness than it is to get permission."
Think that MS's lawyers think that way?!
> sending mail is currently unsupported.
;-)
My company has a similar setup, but we support sending mail. A typical session looks like this:
>~ telnet aol.com smtp
Trying 192.168.10.1...
Connected to mx1.aol.com.
Escape character is '^]'.
220 mx1.aol.com ESMTP Exim 3.35 #1 Thu, 03 Jul 2003 20:27:34 -0400
helo cliftonlabs.com
MAIL FROM: dmartin@foobar.fake
RCPT TO: abuse@aol.com
(and so on.)
Sadly, I'm probably the only slashdotter old enough to remember when you could forge email that way
> I put up with the clunkiness until KDE 3.1 is stable enough for testing.
.debs (they don't upgrade easily, for instance), they work extremely well and I have had no stability issues on them on desktop machines that use woody.
/etc/apt/sources.list line for Adrian Bunk's backport collection is:
Packages for woody are available from download.kde.org. While they aren't nearly as good as official
The other thing that I should have pointed out by now is that there are various backport archives for many common applications, where you can have a stable "core" but run the latest and greatest applications easily. I happen to use Adrian Bunk's backports but there are others too. The
deb http://www.fs.tum.de/~bunk/debian woody/bunk-1 main contrib non-free
YMMV, but I have had very few problems using these backports, and I have used them on many machines.
> So, given Debian's release cycle, that's like, what? One upgrade? Over the past 7 years?
I assume you're being sarcastic, but... There are lots of security updates, so upgrading for those is a fairly regular issue. (The same issues affect RH too, of course.) And given that there's a major release every 18-24 months currently, more frequently in the past, there have been many major upgrades. Note that the machine in question is a server, so not having frequent major upgrades suits me just fine.
On the other hand, my main development box and desktop machine runs the unstable archive, and is usually on the bleeding edge. (Currently it has KDE 3.1.x and mozilla-firebird, g++ 3.3, etc. running on it.)
The reason most jokes are funny is that there is an element of truth to them.