I don't see how this can really help us - I mean, 70-90% of the spam I get is from country codes out side of the US.
70-90% of the spam I get has a faked reply-to address. Are you sure you understand what you are looking at? Besides, goatse.cx isn't really hosted in the Christmas Islands.
Traditionally in the computer industry a White Paper is a high-medium level technical document put out by the marketing department. It generally isn't specifications or the full documentation.
I actually can't see any difference: Copywrited material posted to a message board in contradition to the DMCA. (Forget the stuff about the access controls/zip password).
The only real difference is that Slashdot, newly purchased and high on stock value, probably had a few more dollars and balls in those days. So they told Microsoft to fuck themselves.
And then someone at Microsoft smarter than the lawyers figured that taking on a Unix-oriented board in a legal manner was probably bad PR. (They prefer a more executive-level soft-sell FUD, and one thing that came out of the anti-trust trial is that they laywers are NOT calling the shots there.)
Yup, the US Government, Digital Equipment, and AT+T spent quite a bit of money and effort designing the OSI protocols as their "standard" networking platform. Only to wake up and find that TCP/IP (which was more of an academic garage effort in some corner of the Defense Dept.) was already in use and considered the real standard for hetrogenius networking.
So, OSI was dead on arrival, which is probably why people think it's only a "model" and never a real product. (Time to start looking real close at that MS Exchange Server, boys...)
As for the userfulness of the OSI model, I always thought it's a useful conceptual tool for getting your mind around debugging network systems. Other people seem to agree, which is why you'll find it on page 10 of almost every Networking For Dummies book. Of course, once you've wrapped your mind around the idea of layered protocols, the actual model is worthless.
How much would it cost to hire a paralegal or whatever to sit around and sift through Altavista or Google all day
There were some Scientologists posting on the thread. Apparently they are trained to report this sort of thing back to the mothership so that appropriate action can be taken.
Note that Unisys started collecting on the GIF patent in about 1988 or so. Which means that there's been a dozen or so years for a non-patent encumbered become popular. Which hasn't happened.
There was really two opportunities to solve this:
1) When CompuServe discovered that they fucked up, they should have remidied it by coming out with a GIF2 format. (There must have been a different algorythm than LZW to use...)
2) GIF support should have been built into Mosaic, and never been considered a standard format on the WWW (In fact, did the U of Illinois even have a LZW licence? My guess is not.)
Neither came true, so we wait until the patent expires...
I've taken a look at the WebLogic.java file output from JSPs, and there really isn't anything in there which is glaringly inefficient, except for the fact that every output gets run through some munging method.
But, your point stands that if you are coding for performance, you better fully understand what the JSP engine is doing.
Re:Multi Processors under Win9x
on
Emergence of SMT
·
· Score: 1
http://www.xlr8yourmac.com/G4ZONE/G4_733mhz_review/apple_G4_733_tests.html#storytop has some Single vs. Duel benchmarks on MacOS.
You are right on about the graphic designers. Apple's high end machines are selling almost entirely into nitch markets, and for the most parts those markets use software which takes advantage of multi-processing.
Apple's 'regular user' base is for the most part not buying multi-CPU Macs.
Actually, I thought that productivity did go way up in that period (72-95), while wages went down, which of course led to higher profits. This was one bit reason the stock market was such a good bet from the late 70s until just recently.
(I'm am mostly talking out of my ass, so you might be right on about the "stagnation in productivity", of course.)
Not to mention that two income families effectively doubled the size of the labor pool, which meant that everyone has twice the competition for their position (in theory for higher-end jobs, and in practice for lower-end jobs).
Which is one of the big reasons that real wages didn't increase at all in the US between 1972 and 1995 or so (while the standard of living went way up, primarily due to cheap credit).
This is a cut-and-paste post from about 200 previous slashdot dicussions.
First of all -- You just doesn't understand the huge impact that fragmentation has had on the Unix market and you think you might be able to FUD the facts away by reciting a list.
Second -- I don't care if knowing all the various OSR (.01) releases gives you a stiffy. Keep it to yourself.
Third -- There is a very valid point there about MS's market segmentaiton strategy and the costs businesses bear to support that, but your boring ass post doesn't even scratch the surface.
Great. When you get Fat Ballmer's job, you can make the big decisions in Microsoft's best interest.
Myself, I think they are well aware of potential risks, not just from Free Software, but also from competitors such as IBM/Lotus who have shown willingness to compete with MS on price.
But look at it this way. They've got few problems:
+ Office has so many features they can't even think of new ones to add
+ Customer base can't see value of upgrades
+ Market is saturated, but there's a huge base of users not in licence-compliance
So, they gotta do something. (Being a monopoly is tough, eh?) So their solution is two fold:
+ Add corporate-friendly integration features to Office, making it a client-server platform (should have mentioned this before... See SharePoint portal server and Exchange 2000), increasing both the value of upgrades and the lock-in.
+ Lower the cost of entry and decrease the ease of piracy.
Of course, there's always Plan B, which is what you seem to be suggesting:
+ Put the software in maintence mode and either cut the price to keep marketshare or gouge legacy customers until the end of time.
But they ain't quite there yet, and they won't be until they get new revenue streams that are the size of MS Office's.
You are right on about the Office upgrades. In fact sales of O2K have been pretty slow for Microsoft because most larger corps are sitting on Office 97. Expect sales of Office XP to be even worse, which is why they are moving to copy protection and a subscription sales model (both conspire to squeeze a few nickles out of SOHO users).
I asked someone in the know about the new Microsoft SharePoint portal server, and they didn't seem to think it supported RSS headlines. (Big deal, I wrote one in about 50 lines of code..)
I forgot to ask if it supported Microsoft ActiveChannels (g)
I get 18 at 1:19PM PT. This guy I am replying to posted almost 2 hours after the first person with the correct answer.
God, this place is just crawling with MCSE-wannabes creaming their pants to make some use of the shit they learned at "cram class" while sitting at their useless fucking telephone support jobs.
All we need now is Juan Epstein going OOOOOoooo OOOOOOooo, along with all of these shitheads and this would be the perfect thread. (Actually, if Juan Epstein and cyborg_monkey got into a flamewar, then it would be the perfect thread).
Now, then, the line of defense against needless diversity is to enforce a standard on the metadata, the names, not the values!
I can hear the moans right now: "Ack! Registry!"
The thing to realize is that Unix already uses such a database: It's called the filesystem and the value mappings are called links or symlinks.
It's also a fricking mess, with almost path is hardcoded or a compile-time option. This makes what should be the most basic sysadmin task, moving directories, impossible.
And, sure, Windows is no better. But that's because MS discovered the nice side-effect that piracy is more difficult with hard-coded app paths. Other systems, such as MacOS, do a very nice job of making software relocatable.
This is the standard M$ move... Then, slowly, they will leverage the desktop to work into the server market.
You describe the standard MS playbook. But,.NET in general is not the standard MS playbook.
Ballmer has the government to the left of him, and a seriously depressed stock value to the right of him, and.NET is an attempt to have his cake and eat it too: They can move to cross-platform server space and still "own the platform", as well as leverage the Windows monopoly to get people there.
If they get broken up by the government, or voluntarily split to raise the stock price,.NET allows the "Application" group to divorce itself from the Windows codebase. The problem is, they have just about zero credibility on Unix and with ISPs/ASPs. By doing a soft rollout, they can work the kinks out of.NET on Unix while the Microsoft legion charges ahead and develops on Windows. Then, when the breakup happens (maybe sooner than originally planned!), they've already got a cross-platform product which has been tested and somewhat deployed.
Just add the client pieces, and then it's business as usual - it makes breaking off the application group from the OS side much easier.
but the functionality is there and hogs the application
Well, I can understand not wanting these generic 'handles', but lots of other people do. The functionality to load Word into the IE frame is the exact same as loading the Flash or Java plugin (although a huge security risk!). The functionality to embedding an Excel range into Word is the exact same as embedding a picture or an equation. Can you really get away with not providing extendable applications in this day and age? Can you make the Unix philosophy of small focused tools work on the GUI level (where Unix traditionally has had only monolithic apps) without a real component architecture?
Microsoft has built their software empire on OLE2/COM, and in this sense the free desktop people are playing catchup to Office 4.2 which shipped in 1993 or so. And alas, the functionality is institutionalized in people's use patterns, so it probably does need to be provided for.
But before the free desktop people go whole hog on component embedding, they probably should take a step back and determine why Microsoft chose that route: They did it because they had a monopoly on desktop OSes, and they wanted to entrench that monopoly by providing user tools to entice users to lock up their corporate data *on the desktop*. People run forth and store crucial data in huge Word and Excel files (rather than databases and document systems) because that's the toolset Microsoft gave them.
Perhaps people should take a step back and see where the strengths of Unix are (servers, particularly Internet servers) and what Microsoft is doing to get out of desktop-space and into server-space ("Web Services") before going whole-hog on a client-based component architecture, ala Office 4.2 and Windows 95.
After years of using a Mac before switching to Windows, the thing I miss the most when I go back is the ability to Cut/Copy/Paste files. Maybe it doesn't make "semantic" sense, but if you can use the clipboard to move around all sorts of other objects, I don't see why you can't put files there.
Agreed that the paste-to-text could be handled better.
Judging by the size of Slashdot pasted onto that webslate image, I'm guessing they are proposing to have a ~1024x768 screen in a unit ~7"x5". That's a pretty high density screen, which would be expensive if it's even available.
More realistic in the short term are devices like the Clio, which has a (more realistic) 640x480 screen. But developers could replace the 200Mhz MIPS cpu with a 700Mhz Intel-compatible one, which would probably be not only faster but also have wider software compatiblity (if you go the Windows route).
And one known good way to deter/prevent crime is to not dump hundreds of drunks out on the street at exactly the same time every evening with nowhere for them to go.
You can see this in the US too. Just compare a town where the bars close at 1 AM (you get a bunch of fuckheads standing around the street getting ideas) versus the places that close at a reasonable hour like 3 AM (people gradually get up and go, and the place is almost empty by closing).
This is the same concept as "midnight basketball" in the US. Fucked up punks walking around at night commit crimes, probably a vast majority of 'street crime' - and that includes car thefts, robberies, muggings, and rapes in addition to disorderly conduct. You can prevent the crimes by either preventing people from getting fucked up, or fail that, prevent them from walking around.
Pre-crash there was Atari, Mattel, Coleco, Mangavox/Phillips, Milton Bradley/GCE, Bally, a few others, and some in Japan. Not quite "hundreds" like the guy said, but more than today.
Right: the "American" aspect was the pitch to the Detroit automakers. A quick web check shows that Ford was shipping 8-Track decks starting in 1966, so you're right that my timeline is off.
One correction: Dolby B decks were shipping in 1970, and non-Dolby 'music-grade' cassettes were shipping before then.
Amazon of course went to their investors in about 97-98 and asked them if they should do this. And the answer was "No! We don't care how much money you lose - we want you to become the WalMart of the Internet!".
Now, of course, they have a bunch of money losing properties (online power tool sales?) and they are screwed to the point that WalMart might buy them.
The Compact Cassette format actually predated 8-Track by a number of years.
8-Tracks were invented and sold to Detroit for a couple reasons:
1) It was American technology, where cassettes were invented in the netherlands or something.
2) It was initially a playback-only medium which limited copying, which the RIAA approved of.
There was actually a pretty huge reason to switch from 8-Track to Cassette: 8-Tracks were inherently unreliable.
Slashdot Mirror
I don't see how this can really help us - I mean, 70-90% of the spam I get is from country codes out side of the US.
70-90% of the spam I get has a faked reply-to address. Are you sure you understand what you are looking at? Besides, goatse.cx isn't really hosted in the Christmas Islands.
Traditionally in the computer industry a White Paper is a high-medium level technical document put out by the marketing department. It generally isn't specifications or the full documentation.
I actually can't see any difference: Copywrited material posted to a message board in contradition to the DMCA. (Forget the stuff about the access controls/zip password).
The only real difference is that Slashdot, newly purchased and high on stock value, probably had a few more dollars and balls in those days. So they told Microsoft to fuck themselves.
And then someone at Microsoft smarter than the lawyers figured that taking on a Unix-oriented board in a legal manner was probably bad PR. (They prefer a more executive-level soft-sell FUD, and one thing that came out of the anti-trust trial is that they laywers are NOT calling the shots there.)
Yup, the US Government, Digital Equipment, and AT+T spent quite a bit of money and effort designing the OSI protocols as their "standard" networking platform. Only to wake up and find that TCP/IP (which was more of an academic garage effort in some corner of the Defense Dept.) was already in use and considered the real standard for hetrogenius networking.
So, OSI was dead on arrival, which is probably why people think it's only a "model" and never a real product. (Time to start looking real close at that MS Exchange Server, boys...)
As for the userfulness of the OSI model, I always thought it's a useful conceptual tool for getting your mind around debugging network systems. Other people seem to agree, which is why you'll find it on page 10 of almost every Networking For Dummies book. Of course, once you've wrapped your mind around the idea of layered protocols, the actual model is worthless.
How much would it cost to hire a paralegal or whatever to sit around and sift through Altavista or Google all day
There were some Scientologists posting on the thread. Apparently they are trained to report this sort of thing back to the mothership so that appropriate action can be taken.
Note that Unisys started collecting on the GIF patent in about 1988 or so. Which means that there's been a dozen or so years for a non-patent encumbered become popular. Which hasn't happened.
There was really two opportunities to solve this:
1) When CompuServe discovered that they fucked up, they should have remidied it by coming out with a GIF2 format. (There must have been a different algorythm than LZW to use...)
2) GIF support should have been built into Mosaic, and never been considered a standard format on the WWW (In fact, did the U of Illinois even have a LZW licence? My guess is not.)
Neither came true, so we wait until the patent expires...
I've taken a look at the WebLogic .java file output from JSPs, and there really isn't anything in there which is glaringly inefficient, except for the fact that every output gets run through some munging method.
But, your point stands that if you are coding for performance, you better fully understand what the JSP engine is doing.
http://www.xlr8yourmac.com/G4ZONE/G4_733mhz_review /apple_G4_733_tests.html#storytop has some Single vs. Duel benchmarks on MacOS.
You are right on about the graphic designers. Apple's high end machines are selling almost entirely into nitch markets, and for the most parts those markets use software which takes advantage of multi-processing.
Apple's 'regular user' base is for the most part not buying multi-CPU Macs.
Actually, I thought that productivity did go way up in that period (72-95), while wages went down, which of course led to higher profits. This was one bit reason the stock market was such a good bet from the late 70s until just recently.
(I'm am mostly talking out of my ass, so you might be right on about the "stagnation in productivity", of course.)
Not to mention that two income families effectively doubled the size of the labor pool, which meant that everyone has twice the competition for their position (in theory for higher-end jobs, and in practice for lower-end jobs).
Which is one of the big reasons that real wages didn't increase at all in the US between 1972 and 1995 or so (while the standard of living went way up, primarily due to cheap credit).
This is a cut-and-paste post from about 200 previous slashdot dicussions.
First of all -- You just doesn't understand the huge impact that fragmentation has had on the Unix market and you think you might be able to FUD the facts away by reciting a list.
Second -- I don't care if knowing all the various OSR (.01) releases gives you a stiffy. Keep it to yourself.
Third -- There is a very valid point there about MS's market segmentaiton strategy and the costs businesses bear to support that, but your boring ass post doesn't even scratch the surface.
Great. When you get Fat Ballmer's job, you can make the big decisions in Microsoft's best interest.
Myself, I think they are well aware of potential risks, not just from Free Software, but also from competitors such as IBM/Lotus who have shown willingness to compete with MS on price.
But look at it this way. They've got few problems:
+ Office has so many features they can't even think of new ones to add
+ Customer base can't see value of upgrades
+ Market is saturated, but there's a huge base of users not in licence-compliance
So, they gotta do something. (Being a monopoly is tough, eh?) So their solution is two fold:
+ Add corporate-friendly integration features to Office, making it a client-server platform (should have mentioned this before... See SharePoint portal server and Exchange 2000), increasing both the value of upgrades and the lock-in.
+ Lower the cost of entry and decrease the ease of piracy.
Of course, there's always Plan B, which is what you seem to be suggesting:
+ Put the software in maintence mode and either cut the price to keep marketshare or gouge legacy customers until the end of time.
But they ain't quite there yet, and they won't be until they get new revenue streams that are the size of MS Office's.
100% compatible? I think not.
You are right on about the Office upgrades. In fact sales of O2K have been pretty slow for Microsoft because most larger corps are sitting on Office 97. Expect sales of Office XP to be even worse, which is why they are moving to copy protection and a subscription sales model (both conspire to squeeze a few nickles out of SOHO users).
I asked someone in the know about the new Microsoft SharePoint portal server, and they didn't seem to think it supported RSS headlines. (Big deal, I wrote one in about 50 lines of code..)
I forgot to ask if it supported Microsoft ActiveChannels (g)
How many fucking morons had to point this out?
I get 18 at 1:19PM PT. This guy I am replying to posted almost 2 hours after the first person with the correct answer.
God, this place is just crawling with MCSE-wannabes creaming their pants to make some use of the shit they learned at "cram class" while sitting at their useless fucking telephone support jobs.
All we need now is Juan Epstein going OOOOOoooo OOOOOOooo, along with all of these shitheads and this would be the perfect thread. (Actually, if Juan Epstein and cyborg_monkey got into a flamewar, then it would be the perfect thread).
Now, then, the line of defense against needless diversity is to enforce a standard on the metadata, the names, not the values!
I can hear the moans right now: "Ack! Registry!"
The thing to realize is that Unix already uses such a database: It's called the filesystem and the value mappings are called links or symlinks.
It's also a fricking mess, with almost path is hardcoded or a compile-time option. This makes what should be the most basic sysadmin task, moving directories, impossible.
And, sure, Windows is no better. But that's because MS discovered the nice side-effect that piracy is more difficult with hard-coded app paths. Other systems, such as MacOS, do a very nice job of making software relocatable.
This is the standard M$ move ... Then, slowly, they will leverage the desktop to work into the server market.
.NET in general is not the standard MS playbook.
.NET is an attempt to have his cake and eat it too: They can move to cross-platform server space and still "own the platform", as well as leverage the Windows monopoly to get people there.
.NET allows the "Application" group to divorce itself from the Windows codebase. The problem is, they have just about zero credibility on Unix and with ISPs/ASPs. By doing a soft rollout, they can work the kinks out of .NET on Unix while the Microsoft legion charges ahead and develops on Windows. Then, when the breakup happens (maybe sooner than originally planned!), they've already got a cross-platform product which has been tested and somewhat deployed.
You describe the standard MS playbook. But,
Ballmer has the government to the left of him, and a seriously depressed stock value to the right of him, and
If they get broken up by the government, or voluntarily split to raise the stock price,
Just add the client pieces, and then it's business as usual - it makes breaking off the application group from the OS side much easier.
but the functionality is there and hogs the application
Well, I can understand not wanting these generic 'handles', but lots of other people do. The functionality to load Word into the IE frame is the exact same as loading the Flash or Java plugin (although a huge security risk!). The functionality to embedding an Excel range into Word is the exact same as embedding a picture or an equation. Can you really get away with not providing extendable applications in this day and age? Can you make the Unix philosophy of small focused tools work on the GUI level (where Unix traditionally has had only monolithic apps) without a real component architecture?
Microsoft has built their software empire on OLE2/COM, and in this sense the free desktop people are playing catchup to Office 4.2 which shipped in 1993 or so. And alas, the functionality is institutionalized in people's use patterns, so it probably does need to be provided for.
But before the free desktop people go whole hog on component embedding, they probably should take a step back and determine why Microsoft chose that route: They did it because they had a monopoly on desktop OSes, and they wanted to entrench that monopoly by providing user tools to entice users to lock up their corporate data *on the desktop*. People run forth and store crucial data in huge Word and Excel files (rather than databases and document systems) because that's the toolset Microsoft gave them.
Perhaps people should take a step back and see where the strengths of Unix are (servers, particularly Internet servers) and what Microsoft is doing to get out of desktop-space and into server-space ("Web Services") before going whole-hog on a client-based component architecture, ala Office 4.2 and Windows 95.
After years of using a Mac before switching to Windows, the thing I miss the most when I go back is the ability to Cut/Copy/Paste files. Maybe it doesn't make "semantic" sense, but if you can use the clipboard to move around all sorts of other objects, I don't see why you can't put files there.
Agreed that the paste-to-text could be handled better.
Judging by the size of Slashdot pasted onto that webslate image, I'm guessing they are proposing to have a ~1024x768 screen in a unit ~7"x5". That's a pretty high density screen, which would be expensive if it's even available.
More realistic in the short term are devices like the Clio, which has a (more realistic) 640x480 screen. But developers could replace the 200Mhz MIPS cpu with a 700Mhz Intel-compatible one, which would probably be not only faster but also have wider software compatiblity (if you go the Windows route).
And one known good way to deter/prevent crime is to not dump hundreds of drunks out on the street at exactly the same time every evening with nowhere for them to go.
You can see this in the US too. Just compare a town where the bars close at 1 AM (you get a bunch of fuckheads standing around the street getting ideas) versus the places that close at a reasonable hour like 3 AM (people gradually get up and go, and the place is almost empty by closing).
This is the same concept as "midnight basketball" in the US. Fucked up punks walking around at night commit crimes, probably a vast majority of 'street crime' - and that includes car thefts, robberies, muggings, and rapes in addition to disorderly conduct. You can prevent the crimes by either preventing people from getting fucked up, or fail that, prevent them from walking around.
You mean Post-crash.
Pre-crash there was Atari, Mattel, Coleco, Mangavox/Phillips, Milton Bradley/GCE, Bally, a few others, and some in Japan. Not quite "hundreds" like the guy said, but more than today.
Right: the "American" aspect was the pitch to the Detroit automakers. A quick web check shows that Ford was shipping 8-Track decks starting in 1966, so you're right that my timeline is off.
One correction: Dolby B decks were shipping in 1970, and non-Dolby 'music-grade' cassettes were shipping before then.
Amazon of course went to their investors in about 97-98 and asked them if they should do this. And the answer was "No! We don't care how much money you lose - we want you to become the WalMart of the Internet!".
Now, of course, they have a bunch of money losing properties (online power tool sales?) and they are screwed to the point that WalMart might buy them.
The Compact Cassette format actually predated 8-Track by a number of years.
8-Tracks were invented and sold to Detroit for a couple reasons:
1) It was American technology, where cassettes were invented in the netherlands or something.
2) It was initially a playback-only medium which limited copying, which the RIAA approved of.
There was actually a pretty huge reason to switch from 8-Track to Cassette: 8-Tracks were inherently unreliable.