Microsoft can't drop it entirely. Too many old Windows clients out there only support a given level of SMB and you can't just lock them out from your servers. A DOS 3.3 client can still connect to a Windows 2000 server.
What will no doubt continue to happen is new versions of SMB that are negotiated between machines if both support them. That DOS 3.3 client winds up talking a much simpler version of SMB than a Windows 2000 client would negotiate with a Windows 2000 server. So eventually you discover that your old clients don't support some fancy new feature or performance enhancement. But they should continue to work as they have.
When I worked at Microsoft there were stories that SMB would be dropped in favor of something else. One of these was the famous "HTTP redirector", which would be a client that connected to a web server to replace SMB. One big advantage of HTTP is that it allows a file to be read with one single network roundtrip. SMB separates out the operations of open, read, and close a file (those take up 3 separate SMBs). Now the protocol also allows SMBs to be combined in a single request, but various cruftiness that have built up in the protocol prevents you from actually packaging an open, read, and close in one packet and getting a response back with all the data in one roundtrip. This "small file read" test is a favorite of labs doing performance tests, so although HTTP is slightly larger and slower to parse, that all winds up being noise compared to doing the whole small file read in one roundtrip.
Unfortunately HTTP has some *disadvantages*, in particular a lot of the other semantics supported in SMB, like printing and locks and various security protocols, are not supported, so those would have to be invented. The push for an HTTP redirector seemed to be coming from the Internet Explorer and IIS teams (I worked on NT). In fact one IIS person confidently told me about 4 years ago that SMB was doomed. But it continues to live on and the various HTTP redirector projects seem to have stalled. But for all I know it is still bubbling around somewhere (or several places) at Microsoft.
One thing that I don't think a lot of people realize is how long composing a *good* email can take -- a lot longer than talking on the phone. If you call someone and explain something, they will indicate what parts they understand and what you need to explain more. So it winds up being efficient. Composing a good email is like doing a little presentation, you need to check it over and over to make sure you cover every possible angle. Then you will want to check the spelling, make sure your argument is well stated, and so on, because you only get one chance to get it right (of course, once you get it right, the mail can be forwarded around without the "Telephone" effect, the gradual entropization of ear-to-mouth communication).
The same applies to Slashdot posts also!
- adam
P.S. Wait a minute, that was a Jon Katz article that was topical, insightful, well-argued...what is going on here?!?!?
not disclosing patents to standard bodies?
on
MS getting rid of SAMBA?
·
· Score: 5, Informative
Most parties with patents are not required to disclose them in the standards setting process, van den Beld said.
Is this true? Consider this article from Fortune about
Rambus, in which they were dinged for not disclosing a patent to a standards committee.
The article mentions that Sun and Dell got in trouble for similar things, and had to license the patents royalty-free. Dell had a patent on VL-BUS technology, and Sun had one on DRAMs for SparcStations that Kingston complained about. The Dell story (from 1996) is summarized
here
and this is from the
FTC,
while the Sun case (from this year) is mentioned
here
and
here.
If Bill Gates tells some people at Microsoft, "No you can't pursue that project," then they are basically done with it. They won't get the resources to work on it, they won't be able to check in changes, they might not even get access to the source code they need. Sometimes skunkworks projects are shielded by sympathetic VPs but it is rare.
Meanwhile if some Linux people want to do something unusual, they are free to do it, then present it to the rest of the community when they see fit. No need to justify it ahead of time or battle out with other projects needing resources. Of course their work may still be rejected but they can get a much better shot. In a sense this can lead to more work "waste" but also more avenues are explored.
two quotes from the book
on
Breaking Windows
·
· Score: 2, Insightful
"Gates was an introvert who liked people mainly for the intellectual simulation they provided...Ballmer was an extrovert who enjoyed the pleasures of socializing for his own sake."
"Ballmer's easy bonhomie and meat-and-potatoes approach to the business seemed to be just what Microsoft needed, the perfect antidote to Gates's enigmatic aloofness."
So I'm sure Bank would say (and I say), "Go Steve!" He does get the employees fired up...and it's nice to see a guy worth $25 billion (or whatever) who is running a $30 billion company but still doesn't take himself all that seriously.
- adam
Re:Some other books
on
Breaking Windows
·
· Score: 3, Informative
Microserfs is a good read but only the first chapter or so is really about Microsoft.
Renegades is interesting but mostly just story-telling. I mean jeez he is talking about three *evangelists* and he never really picks up on how significant their position is within Microsoft -- instead he talks about how zany they were (and they were!)
Showstopper I'm not too excited about. First of all I don't think the author really understood how software works. His analogies are execrable. Plus he makes it sound like the whole thing was written by about 15 people and they were all freaks. And he never mentioned me.
That's why I liked "Breaking Windows", it was the first book about Microsoft (besides mine, natch) where I did not spend some significant period of time shaking my head in disbelief.
- adam
Re:Its interesting that Internal...
on
Breaking Windows
·
· Score: 1
This is true. One case is discussed in the book, a company called CrossGain. After some legal wrangling, the Microsoft expats all had to quit, then rejoin the company when their one year was up.
It was a little of both. I did mostly work on loaders, kernels, and various device drivers, thus "low-level code." I also didn't rise very far up the management chain, when I left I only had 2 people working for me.
However, Microsoft has a system in place to promote people without making them managers, so I didn't just stagnate for ten years.
- adam
P.S. I was lurking in this thread (see discussion
here)
but finally decided to nuke my moderations and post in the thread instead.
I feel your pain. However, as I have unfortunately discovered, even a/. front-page listing won't do much to help a print-on-demand book get traction in the literary world.
- adam
P.S. Email me if you want my longer take on the subject.
I don't mean to have the a-accent-grave in French be mapped to a plain English a. Certainly you should keep the accents. But do the g and r and c and e need to be different characters?
Mapping these things is a nightmare. Imagine someone writing say a BIOS, or something else with limited storage and code, who wants to display a startup message...do they need to store a separate Unicode message for each language just so they can say "Dell Computer" or whatever properly for each one? And then you have to store all those glyphs for multiple fonts, so you would probably wind up mapping them all back on top of each other. Gack.
The purists who want 4-byte characters go beyond just wanting to allow 50,000 Kanji or insisting that Japanese and Chinese Kanji with the same stroke pattern not share the same character. They want a separate character for the English lower-case 'e', the French lower-case 'e', the German lower-case 'e', etc. This is not at all necessary. YES, there may be some Kanji that fall out of use if the set listed in the Unicode standard becomes the only one used, but you have to counter that with the fact that suddenly these languages can have a universally-recognized way to encode them, as opposed to the 5 of whatever ways that previously existed to encode Japanese (which all had limited character sets anyway).
UTF-8 is very nice because 7-bit characters encode as one byte. Also it is defined so there won't be a NULL or a hex 01B (decimal 27 -- the telnet escape character) anywhere in the data stream, even in the second or third byte of an encoded character. So it will generally be passed through correctly by programs expecting straight 8-bit ASCII. UTF-8 is also encoded and decoded via a trivial algorithm, as opposed to the DBCS used in Windows which needs lookup tables.
One negative of UTF-8 is that Unicode characters at 0x8000 or above (using more than 11 bits) encode in UTF-8 as 3 bytes, not 2 as in Unicode. I think that range includes things like Arabic and some Indian written languages. But I think that tradeoff is worth it.
Statements like "Microsoft typically links its software and operating systems (SQLServer won't run outside an MS environment; Office is only ported to the Mac)" and "With HailStorm, Microsoft has abandoned tying its major software offerings to its client operating systems" miss the point.
Don't think "operating system". Think "platform"..NET is the new platform, and Microsoft is surely going to tie its apps to a single platform. It's just that the platform now is more than just an OS.
You said
"neither of your parents went to a French school in their youth" you had to go to French school...what you meant was if neither of your parents went to ENGLISH school.
It used to be if they went to English school anywhere in the world, it was OK...I think it was changed to it had to be within Canada or you were stuck with French for your kids.
- adam...who went to English school in Quebec, so my kids can go to English school also--which would matter if I had any intention of raising them in Quebec
I wrote a two-part article about this on osopinion.com last December (here are
part 1
and
part 2).
- adam
Nice for the 0.1% of the world it applies to
on
The Hacker Ethic
·
· Score: 2
"But offices, factories and retail stores aren't going away."
Gee, ya think?!?
To take advantage of this allegedly new ethic, you need a job with certain very specific characteristics. Little direct interaction with customers, not tied to a specific physical location, and requiring a particular intellectual aptitude that is uncommon. So it's great that people like this are now being spoiled even more by their employers. But the notion that this is replacing the Protestant work ethic is so absurd that it's not even funny.
Try telling this to someone working in a sweatshop in a third-world country, or someone working three jobs to get by in America, or anyone who needs a union to protect their job. "It's not society's fault...it's your fault...you need a new work ethic!! That's all!" Yeah right.
In any case, this whole idea is not new, since for centuries it has been the standard way of working among academic researchers, who were among the few who had jobs that actually matched the criteria.
I have to applaud your open-mindedness after years of Microsoft bashing...I heartily agree, since a lot of what you say in this essay is what I say in
chapter 13
of my
book.
- adam
This is one of the first mainstream articles I have seen that is consistent with my feeling about what.NET will be [sound of my own horn being tooted], which I described in an
article
on OsOpinion.com last November.
- adam
What .NET really is...
on
Perl and .NET
·
· Score: 1
It's not just a reply to Java, or a way to do subscription-based applications, or whatever. It's Microsoft's way to define the Internet as a platform for applications and "drivers", the same way they did for Windows (drivers in this case are really service providers).
If anyone cares, I wrote an
article for
OsOpinion.com which gives my view on all this.
"this is a short-sighted betrayal for millions of mostly younger people who've learned to love music online, and who spend billions of dollars on it."
That's a nicely constructed sentence, implying that people have spent billions of dollars on online music -- when in fact the whole point of Napster is that people have spent exactly zero dollars on the music.
Or this one:
"The ruling will definitely set the tone for how intellectual property is defined on the Internet"
You mean set the tone that artists actually have some right not to have their work ripped off for free?
The argument that Napster helps sales is a big red herring. Maybe it does -- but artists should have a right to decide if they want to use Napster as a loss leader or not. If you tell me that smacking me in the head will make me feel better, well maybe it will, but can I decide if I want you to?
The thing that really peeves me about Napster is that it is being presented as a halfway step towards real secure digital content. Yeah right. Napster may be a cool hack, but all it really is is a database and a file transfer protocol -- woohoo, weren't those problems solved about 25 years ago? Napster conveniently avoids any hard issues involved. If secure digital content is solved, it will be through combined efforts of content producers, hardware companies, software companies, etc. Napster's only effect will be to spur the process along; in no way does it deserve a seat at the table.
Slashdot Mirror
What will no doubt continue to happen is new versions of SMB that are negotiated between machines if both support them. That DOS 3.3 client winds up talking a much simpler version of SMB than a Windows 2000 client would negotiate with a Windows 2000 server. So eventually you discover that your old clients don't support some fancy new feature or performance enhancement. But they should continue to work as they have.
When I worked at Microsoft there were stories that SMB would be dropped in favor of something else. One of these was the famous "HTTP redirector", which would be a client that connected to a web server to replace SMB. One big advantage of HTTP is that it allows a file to be read with one single network roundtrip. SMB separates out the operations of open, read, and close a file (those take up 3 separate SMBs). Now the protocol also allows SMBs to be combined in a single request, but various cruftiness that have built up in the protocol prevents you from actually packaging an open, read, and close in one packet and getting a response back with all the data in one roundtrip. This "small file read" test is a favorite of labs doing performance tests, so although HTTP is slightly larger and slower to parse, that all winds up being noise compared to doing the whole small file read in one roundtrip.
Unfortunately HTTP has some *disadvantages*, in particular a lot of the other semantics supported in SMB, like printing and locks and various security protocols, are not supported, so those would have to be invented. The push for an HTTP redirector seemed to be coming from the Internet Explorer and IIS teams (I worked on NT). In fact one IIS person confidently told me about 4 years ago that SMB was doomed. But it continues to live on and the various HTTP redirector projects seem to have stalled. But for all I know it is still bubbling around somewhere (or several places) at Microsoft.
- adam
The same applies to Slashdot posts also!
- adam
P.S. Wait a minute, that was a Jon Katz article that was topical, insightful, well-argued...what is going on here?!?!?
Is this true? Consider this article from Fortune about Rambus, in which they were dinged for not disclosing a patent to a standards committee.
The article mentions that Sun and Dell got in trouble for similar things, and had to license the patents royalty-free. Dell had a patent on VL-BUS technology, and Sun had one on DRAMs for SparcStations that Kingston complained about. The Dell story (from 1996) is summarized here and this is from the FTC, while the Sun case (from this year) is mentioned here and here.
- adam
- adam
P.S. 100th post!
- adam
Meanwhile if some Linux people want to do something unusual, they are free to do it, then present it to the rest of the community when they see fit. No need to justify it ahead of time or battle out with other projects needing resources. Of course their work may still be rejected but they can get a much better shot. In a sense this can lead to more work "waste" but also more avenues are explored.
- adam
Who told you I got fired?
- adam
"Ballmer's easy bonhomie and meat-and-potatoes approach to the business seemed to be just what Microsoft needed, the perfect antidote to Gates's enigmatic aloofness."
So I'm sure Bank would say (and I say), "Go Steve!" He does get the employees fired up...and it's nice to see a guy worth $25 billion (or whatever) who is running a $30 billion company but still doesn't take himself all that seriously.
- adam
Renegades is interesting but mostly just story-telling. I mean jeez he is talking about three *evangelists* and he never really picks up on how significant their position is within Microsoft -- instead he talks about how zany they were (and they were!)
Showstopper I'm not too excited about. First of all I don't think the author really understood how software works. His analogies are execrable. Plus he makes it sound like the whole thing was written by about 15 people and they were all freaks. And he never mentioned me.
That's why I liked "Breaking Windows", it was the first book about Microsoft (besides mine, natch) where I did not spend some significant period of time shaking my head in disbelief.
- adam
- adam
However, Microsoft has a system in place to promote people without making them managers, so I didn't just stagnate for ten years.
- adam
P.S. I was lurking in this thread (see discussion here) but finally decided to nuke my moderations and post in the thread instead.
- adam
P.S. Email me if you want my longer take on the subject.
I don't mean to have the a-accent-grave in French be mapped to a plain English a. Certainly you should keep the accents. But do the g and r and c and e need to be different characters?
Mapping these things is a nightmare. Imagine someone writing say a BIOS, or something else with limited storage and code, who wants to display a startup message...do they need to store a separate Unicode message for each language just so they can say "Dell Computer" or whatever properly for each one? And then you have to store all those glyphs for multiple fonts, so you would probably wind up mapping them all back on top of each other. Gack.
- adam
UTF-8 is very nice because 7-bit characters encode as one byte. Also it is defined so there won't be a NULL or a hex 01B (decimal 27 -- the telnet escape character) anywhere in the data stream, even in the second or third byte of an encoded character. So it will generally be passed through correctly by programs expecting straight 8-bit ASCII. UTF-8 is also encoded and decoded via a trivial algorithm, as opposed to the DBCS used in Windows which needs lookup tables.
One negative of UTF-8 is that Unicode characters at 0x8000 or above (using more than 11 bits) encode in UTF-8 as 3 bytes, not 2 as in Unicode. I think that range includes things like Arabic and some Indian written languages. But I think that tradeoff is worth it.
- adam
Don't think "operating system". Think "platform". .NET is the new platform, and Microsoft is surely going to tie its apps to a single platform. It's just that the platform now is more than just an OS.
- adam
You said "neither of your parents went to a French school in their youth" you had to go to French school...what you meant was if neither of your parents went to ENGLISH school. It used to be if they went to English school anywhere in the world, it was OK...I think it was changed to it had to be within Canada or you were stuck with French for your kids. - adam...who went to English school in Quebec, so my kids can go to English school also--which would matter if I had any intention of raising them in Quebec
I wrote a two-part article about this on osopinion.com last December (here are part 1 and part 2). - adam
"But offices, factories and retail stores aren't going away."
Gee, ya think?!?
To take advantage of this allegedly new ethic, you need a job with certain very specific characteristics. Little direct interaction with customers, not tied to a specific physical location, and requiring a particular intellectual aptitude that is uncommon. So it's great that people like this are now being spoiled even more by their employers. But the notion that this is replacing the Protestant work ethic is so absurd that it's not even funny.
Try telling this to someone working in a sweatshop in a third-world country, or someone working three jobs to get by in America, or anyone who needs a union to protect their job. "It's not society's fault...it's your fault...you need a new work ethic!! That's all!" Yeah right.
In any case, this whole idea is not new, since for centuries it has been the standard way of working among academic researchers, who were among the few who had jobs that actually matched the criteria.
- adam
I have to applaud your open-mindedness after years of Microsoft bashing...I heartily agree, since a lot of what you say in this essay is what I say in chapter 13 of my book. - adam
- adam
If anyone cares, I wrote an article for OsOpinion.com which gives my view on all this.
- adam
"this is a short-sighted betrayal for millions of mostly younger people who've learned to love music online, and who spend billions of dollars on it."
That's a nicely constructed sentence, implying that people have spent billions of dollars on online music -- when in fact the whole point of Napster is that people have spent exactly zero dollars on the music.
Or this one:
"The ruling will definitely set the tone for how intellectual property is defined on the Internet"
You mean set the tone that artists actually have some right not to have their work ripped off for free?
The argument that Napster helps sales is a big red herring. Maybe it does -- but artists should have a right to decide if they want to use Napster as a loss leader or not. If you tell me that smacking me in the head will make me feel better, well maybe it will, but can I decide if I want you to?
The thing that really peeves me about Napster is that it is being presented as a halfway step towards real secure digital content. Yeah right. Napster may be a cool hack, but all it really is is a database and a file transfer protocol -- woohoo, weren't those problems solved about 25 years ago? Napster conveniently avoids any hard issues involved. If secure digital content is solved, it will be through combined efforts of content producers, hardware companies, software companies, etc. Napster's only effect will be to spur the process along; in no way does it deserve a seat at the table.
- adam