I think it was in his Byte column where he wrote of his experiences trying to get started developing for OS/2, and Win95.
At a trade show, he went to the Microsoft booth, and asked what he had to do to get started with Windows development. They handed him a developers kit right there.
He went to the IBM booth, and asked then what he had to do to get started with OS/2 development. They handed him an application to their developer program so he could ask for permission to develop for OS/2 (for a large fee, of course).
I realized OS/2 was truly doomed about a year later, when I went into Egghead, and saw MSDN Library subscriptions for sale. The only OS/2 development tool I saw at Egghead was the Watcom C/C++ compiler.
Another thing that hurt OS/2 was the lack of good third-party documentation. Where was the equivalent of Petzold's wonderful Windows books, that got so many of us started on Windows programming? There IS a book on OS/2 programming by Petzold, but it was often out of print. I'm sure IBM could have managed to get it back into print if they'd wanted.
As many others have already pointed out, many cable companies, including some of the biggest, offer and support PVRs, so they clearly don't hate the concept.
It's the ReplayTV that they hate. Take a look at its feature list, and it is pretty clear that it is designed to aid people spreading copies of programs around.
I wouldn't say that ReplayTV is trying to be the Napster of video, though. Napster did have substantial non-piracy uses, such as independent bands using it to promote their music. It's hard to even pretend that ReplayTV's video sharing features will have any significant non-infringing use.
The don't play on Linux with Xine, with whatever codecs come with it.
On my XP system, WMP 8 gets an error trying to download the necessary codecs. I downloaded the latest codec package from MS and installed them all, and they still won't play.
The media player on Win98 also doesn't come with the right codec.
When making a video to show off your stuff, wouldn't it be a good idea to actually use a widely available format?
Well, it's also umount is sh, ash, csh, tcsh, and pretty much every other shell. Why would you think the shell makes a difference? Almost all shells get the name from the name of the command itself.
It is not unreasonable for employers to not want their support people providing help in public forums while identified as working for the company.
People will come to expect support via that forum. What happens when the employee tires of spending his own time providing support there? He stops, and then the company gets a reputation in the forum of not providing good tech support!
If the employee just wants to be nice, and help people out that they see having trouble in a forum they are reading for their own enjoyment, they should simply not mention where they work. Help out as an unusually knowledgable consumer.
I wouldn't say Clarke predicted geosynchronous satellites. I'd say he invented them.
Data oriented computing?
on
More on Longhorn
·
· Score: 3, Interesting
The "one program" thing reminds me of early Mac, in a way. I don't mean in implementation, but in feel.
I first noticed this back in 1985, when I bought my first Mac. On my Unix systems, the mindset is "I'm using programs to manipulate data". E.g., to change a file, I run vi (vim nowadays), and tell the program what changes I want made to the file, and the program makes the changes.
On a Mac, it feels (or, rather, it used to...Apple has moved away from this) like I'm directly making the changes. I wasn't telling MacWrite to change my file--I was changing my file, using MacWrite as my tool.
I think it felt this way because of the interface consistency among most programs. MacWrite might provide more editing options than, say, a paint program's text tool, but they were consistent. This made the programs feel like they were just part of the computer, rather than the focus of the computer like they are in Unix and Windows. The WYSIWYG aspect also certainly helped a lot.
I think that this is what Microsoft is talking about when they say one program to do everything. I doubt they mean one giant monolithic uberprogram to do everything--they've spent years moving everything in site to be collections of components, and I don't think they'll abandon that approach.
SIMs as experiment
on
Virtual Simerica
·
· Score: 5, Interesting
There was a very interesting story in Analog about a year or two ago. The story started out with the protaganist attending two conferences that were taking place in the same city. One was a conference on virtual reality. The other was a conference on nanotechnology.
The protagonist met an interesting woman at the nanotech conference. The next day, he met a woman who could almost be her twin, but not quite, at the VR conference.
He managed to figure out that the woman from the nanotech conference was there to kill the leading nanotech researcher, and the woman from the VR conference was there to kill the leading VR researcher.
It turns out that both women were from the future...but very different futures. In one, nanotech had been developed, but fell into the wrong hands. The world was under the power of a dictator, whose nanotech made him pretty much invincible. In the other, VR had been developed to the point that virtual worlds had become more interesting to many people than the real world. People were "living" in VR instead of reality. As a side effect of this, people had been able to experiment with different social structures, and they had figured out how to basically implement Utopia--but because so many people had slacked off from real life to do this, the infrastructure was collapsing, and so mankind was doomed.
The protagonist realized that VR-world went bad because nanotech had not been developed in that timeline--because someone had assisinated the lead nanotech researcher! In nanotech-world, the dictator had been able to take over because society had not been restructured along the lines discovered in VR-world, because VR had not been developed, because someone had killed the leading VR researcher. If both VR and nanotech were developed, things would have been great.
Since they're throwing the word theft around so lightly, can I accuse such sites of theft simply because they're stealing my metered bandwidth by forcing me to watch their bloated animated ads?
Not unless you can explain how they forced you to go to the sites in the first place.
This is probably the ultimate MMORPG license. Unless someone comes out with one based on Monty Python and the Holy Grail, I'm sure Sony and LucasArts will have cornered the market
How about Star Trek? There's probably room for an MMORPG there.
Some classic SF could be interesting. For example, Asimov's Foundation series, or Niven's Known Space series. The ringworld alone has room for a rather large game.
There's also alternate history, which is what Mythic is doing with their next MMORPG. That one will be set in a galaxy where the Roman Empire survived, and became a galactic empire.
I'm strangely unexcited by SWG. For me, Star Wars is something to watch, not something to do. It's only really interesting to me in that they are supposedly using much of the same engine the EQ2 will use, so as an EQ player, I'm curious to see it.
The current nvidia drivers cause an instant reboot when I try to start X. The most interesting thing about this is that on my other machine, which is my gaming machine and runs WinXP, the only blue screens I get are in Everquest and Dark Age of Camelot, and I suspect that those are due to video driver problems--and that machine has an NVidia card, too.
It's nice to see that NVidia's unified driver architecture works--they can take down both my Linux and WinXP machine!
Say...why not combine this with a wireless keyboard and wireless mouse! Then you could compute anywhere in your house! And hey...what if we put a small hard disk in, and added a CPU...then we wouldn't need the wireless connection to the PC!
DS9 was a very good Trek...but they missed a great opportunity at the end. Consider these facts:
The Changelings are worried enough about the Federation to engage in a major war against it
Changelings are excellent spies. Even after the Federation was aware of them, and taking measures against them, we saw Changelines take the place of the top Klingon, and infiltrate very high into Starfleet.
Starfleet has been penetrated before. E.g., in TNG, there were those aliens that crawled down your throat and took over...those got into, I think, at least an Admiral.
We've seen many more, in TNG and the OS, that tried to penetrate Starfleet, but were foiled.
We also saw in several DS9 episodes that there are corrupt or rogue elements within Starfleet. Section 31, for example. Whether these are really due to bad humans, or the result of alien penetration is up for speculation.
Given all that, here's how they could have ended the series. The Changelings are fighting the Federation not because they just have an irrational hatred of solids from years of mistreatment before they started the Dominion. Rather, they oppose the Federation because they have spies at the highest levels and know the truth. The "good guy" Federation that Sisko is fighting for is just a sham, covering up evil people (or aliens) that are trying to grab galaxy-wide power.
The series could have ended with Sisko realizing this, and changing sides, working as a rebel trying to overthrow the evil side of the Federation to save the good side.
The right way to do things is for the person who makes the release package (e.g., the tarball, or the rpm, or whatever) to digitally sign it. They should do the signing on a machine other than the web server or FTP server. Ideally, they do the signing on their development machine, which is safetly tucked away on a network that crackers can't get to.
The funny thing about the paranoids who build from source is that, unless they actually look at the source, it doesn't gain them anything. There are three ways to build from source.
1. Just grab the source and build it. This is no better than grabbing a binary and running it, as far as security goes.
2. Grab the source, check the MD5 sum, and then build it. This is no better than grabbing the binary, checking the binary's MD5 sum, and then running it.
3. Grab the source, diff it against the previous source you were running, and at least glance at the diffs to see if anything looks suspicious. This is the only way that using source gives you more security than using the binary.
People using source for security who are in category 1 or 2 are just fooling themselves.
IV, V, and VI are like the Mac, and I, II, and likely III are like the Amiga. The Mac interface was better than the Amiga interface from a visual point of view, because the Amiga people were so happy and in love with their fancy colors that they went way overboard. Apple, when it got color, used it tastefully and only when appropriate, and so looked much better.
In I, II, and likely III, the effects and the components that make up individual scenes are way better than anything in IV, V, and VI, but the overall visual effect in IV, V, and VI is better.
The victim is just as dead whether or not it was premeditated. The point is that we don't just look at the action, but the mental state behind the action.
BTW, I think the original premise that the only difference between a hate crime and a regular crime is the thoughts of the perpetrator is wrong. It is a different act, because of the affect on other people. A hate crime is both a crime against the direct victim, and an act of terrorism against a whole group, designed to cause fear and intimidation among that group. That justifies additional punishment.
Hate "crimes" are inherently though crimes. They punish you additionally for what you think, rather than only based on what you do
So? Most crimes take into account the mental state of the perpetrator. E.g., consider in most jurisdictions the difference between first degree murder, second degree murder, and manslaughter. Would you say that first degree murder is a thought crime?
I think a big part of it is that quite a few OS X apps do things in a way that makes them seem slower than they are.
Take IE, for example. It seems to wait to display the page until it has the whole thing ready to render. On a big slashdot story, that can take a while. Compare to, say, most browsers on Linux, which seem to display while the page is still downloading. Browsing seems way faster on my home system on a 144 Kbit/second connection with Linux than it does at work on OS X on a T3.
On the other hand, I do have evidence that the Mac is actually slow. E.g., when I start to load a slashdot page at work, I often give up, switch over to the XP machine on the KVM switch, and go load it there, and finish ahead of the Mac. The XP machine is an ancient P2 400 with 384 megs of RAM, the Mac is an ancient B&W G3 300 MHz with about 600 meg of RAM, so the machines are comparable (both pathetic by modern standards, but comparable). So, it actually appears that the Mac is slow at browsing, and IE works in such a way to emphasize that slowness, making it seem unbearably slow.
Also, a lot of apps, and Finder, aren't as threaded as they could be. While IE, for instance, is busy getting that big slashdot page ready to display, the dreaded spinning color-ball shows up, so you can't switch back and view the other pages you were reading.
Finally, much Apple software IS slow. There's a thread on comp.sys.mac.advocacy about this right now, where someone was saying that the new generation of iApps seem slower than the previous iApps, and pointing out an apparent correlation between those written in Carbon (fast) and Cocoa (slow). However, other people have pointed out examples of fast Cocoa apps, so that is not the problem. Most interesting was someone who wrote their own photo manager, and compared to iPhoto. For some things, his is 2 orders of magnitude faster than iPhoto. Evidently, Apple simply used crappy algorithms in iPhoto. Apple's mail program is similarly problematic when mailboxes get large. A lot of people on comp.sys.mac.advocacy have given up on it and switched to Eudora, and report their Macs are nice and fast at mail then.
Slashdot Mirror
At a trade show, he went to the Microsoft booth, and asked what he had to do to get started with Windows development. They handed him a developers kit right there.
He went to the IBM booth, and asked then what he had to do to get started with OS/2 development. They handed him an application to their developer program so he could ask for permission to develop for OS/2 (for a large fee, of course).
I realized OS/2 was truly doomed about a year later, when I went into Egghead, and saw MSDN Library subscriptions for sale. The only OS/2 development tool I saw at Egghead was the Watcom C/C++ compiler.
Another thing that hurt OS/2 was the lack of good third-party documentation. Where was the equivalent of Petzold's wonderful Windows books, that got so many of us started on Windows programming? There IS a book on OS/2 programming by Petzold, but it was often out of print. I'm sure IBM could have managed to get it back into print if they'd wanted.
It's the ReplayTV that they hate. Take a look at its feature list, and it is pretty clear that it is designed to aid people spreading copies of programs around.
I wouldn't say that ReplayTV is trying to be the Napster of video, though. Napster did have substantial non-piracy uses, such as independent bands using it to promote their music. It's hard to even pretend that ReplayTV's video sharing features will have any significant non-infringing use.
On my XP system, WMP 8 gets an error trying to download the necessary codecs. I downloaded the latest codec package from MS and installed them all, and they still won't play.
The media player on Win98 also doesn't come with the right codec.
When making a video to show off your stuff, wouldn't it be a good idea to actually use a widely available format?
Well, it's also umount is sh, ash, csh, tcsh, and pretty much every other shell. Why would you think the shell makes a difference? Almost all shells get the name from the name of the command itself.
- People will come to expect support via that forum. What happens when the employee tires of spending his own time providing support there? He stops, and then the company gets a reputation in the forum of not providing good tech support!
If the employee just wants to be nice, and help people out that they see having trouble in a forum they are reading for their own enjoyment, they should simply not mention where they work. Help out as an unusually knowledgable consumer.He didn't say that they were doing RAID-5 on the video editing systems. The RAID-5 was on the file server.
I wouldn't say Clarke predicted geosynchronous satellites. I'd say he invented them.
I first noticed this back in 1985, when I bought my first Mac. On my Unix systems, the mindset is "I'm using programs to manipulate data". E.g., to change a file, I run vi (vim nowadays), and tell the program what changes I want made to the file, and the program makes the changes.
On a Mac, it feels (or, rather, it used to...Apple has moved away from this) like I'm directly making the changes. I wasn't telling MacWrite to change my file--I was changing my file, using MacWrite as my tool.
I think it felt this way because of the interface consistency among most programs. MacWrite might provide more editing options than, say, a paint program's text tool, but they were consistent. This made the programs feel like they were just part of the computer, rather than the focus of the computer like they are in Unix and Windows. The WYSIWYG aspect also certainly helped a lot.
I think that this is what Microsoft is talking about when they say one program to do everything. I doubt they mean one giant monolithic uberprogram to do everything--they've spent years moving everything in site to be collections of components, and I don't think they'll abandon that approach.
The protagonist met an interesting woman at the nanotech conference. The next day, he met a woman who could almost be her twin, but not quite, at the VR conference.
He managed to figure out that the woman from the nanotech conference was there to kill the leading nanotech researcher, and the woman from the VR conference was there to kill the leading VR researcher.
It turns out that both women were from the future...but very different futures. In one, nanotech had been developed, but fell into the wrong hands. The world was under the power of a dictator, whose nanotech made him pretty much invincible. In the other, VR had been developed to the point that virtual worlds had become more interesting to many people than the real world. People were "living" in VR instead of reality. As a side effect of this, people had been able to experiment with different social structures, and they had figured out how to basically implement Utopia--but because so many people had slacked off from real life to do this, the infrastructure was collapsing, and so mankind was doomed.
The protagonist realized that VR-world went bad because nanotech had not been developed in that timeline--because someone had assisinated the lead nanotech researcher! In nanotech-world, the dictator had been able to take over because society had not been restructured along the lines discovered in VR-world, because VR had not been developed, because someone had killed the leading VR researcher. If both VR and nanotech were developed, things would have been great.
It was a pretty cool story.
Not unless you can explain how they forced you to go to the sites in the first place.
That might be relevant if the article was about Verizon cell phone service. However, the article is about Verizon local phone service.
How about Star Trek? There's probably room for an MMORPG there.
Some classic SF could be interesting. For example, Asimov's Foundation series, or Niven's Known Space series. The ringworld alone has room for a rather large game.
There's also alternate history, which is what Mythic is doing with their next MMORPG. That one will be set in a galaxy where the Roman Empire survived, and became a galactic empire.
I'm strangely unexcited by SWG. For me, Star Wars is something to watch, not something to do. It's only really interesting to me in that they are supposedly using much of the same engine the EQ2 will use, so as an EQ player, I'm curious to see it.
It's nice to see that NVidia's unified driver architecture works--they can take down both my Linux and WinXP machine!
Foul language is a banable offense on most online games.
Say...why not combine this with a wireless keyboard and wireless mouse! Then you could compute anywhere in your house! And hey...what if we put a small hard disk in, and added a CPU...then we wouldn't need the wireless connection to the PC!
The Changelings are worried enough about the Federation to engage in a major war against it
Changelings are excellent spies. Even after the Federation was aware of them, and taking measures against them, we saw Changelines take the place of the top Klingon, and infiltrate very high into Starfleet.
Starfleet has been penetrated before. E.g., in TNG, there were those aliens that crawled down your throat and took over...those got into, I think, at least an Admiral.
We've seen many more, in TNG and the OS, that tried to penetrate Starfleet, but were foiled.
We also saw in several DS9 episodes that there are corrupt or rogue elements within Starfleet. Section 31, for example. Whether these are really due to bad humans, or the result of alien penetration is up for speculation.
Given all that, here's how they could have ended the series. The Changelings are fighting the Federation not because they just have an irrational hatred of solids from years of mistreatment before they started the Dominion. Rather, they oppose the Federation because they have spies at the highest levels and know the truth. The "good guy" Federation that Sisko is fighting for is just a sham, covering up evil people (or aliens) that are trying to grab galaxy-wide power.
The series could have ended with Sisko realizing this, and changing sides, working as a rebel trying to overthrow the evil side of the Federation to save the good side.
Don't forget the Ferengi. They were developed quite a bit in DS9.
But is it normally a different server running a different operating system and a different web/ftp server, administered by different people?
If not, then there's no reason to trust it any more than the the server with the trojan.
The right way to do things is for the person who makes the release package (e.g., the tarball, or the rpm, or whatever) to digitally sign it. They should do the signing on a machine other than the web server or FTP server. Ideally, they do the signing on their development machine, which is safetly tucked away on a network that crackers can't get to.
1. Just grab the source and build it. This is no better than grabbing a binary and running it, as far as security goes.
2. Grab the source, check the MD5 sum, and then build it. This is no better than grabbing the binary, checking the binary's MD5 sum, and then running it.
3. Grab the source, diff it against the previous source you were running, and at least glance at the diffs to see if anything looks suspicious. This is the only way that using source gives you more security than using the binary.
People using source for security who are in category 1 or 2 are just fooling themselves.
In I, II, and likely III, the effects and the components that make up individual scenes are way better than anything in IV, V, and VI, but the overall visual effect in IV, V, and VI is better.
That was a great show. Ickus rules!
BTW, I think the original premise that the only difference between a hate crime and a regular crime is the thoughts of the perpetrator is wrong. It is a different act, because of the affect on other people. A hate crime is both a crime against the direct victim, and an act of terrorism against a whole group, designed to cause fear and intimidation among that group. That justifies additional punishment.
So? Most crimes take into account the mental state of the perpetrator. E.g., consider in most jurisdictions the difference between first degree murder, second degree murder, and manslaughter. Would you say that first degree murder is a thought crime?
Take IE, for example. It seems to wait to display the page until it has the whole thing ready to render. On a big slashdot story, that can take a while. Compare to, say, most browsers on Linux, which seem to display while the page is still downloading. Browsing seems way faster on my home system on a 144 Kbit/second connection with Linux than it does at work on OS X on a T3.
On the other hand, I do have evidence that the Mac is actually slow. E.g., when I start to load a slashdot page at work, I often give up, switch over to the XP machine on the KVM switch, and go load it there, and finish ahead of the Mac. The XP machine is an ancient P2 400 with 384 megs of RAM, the Mac is an ancient B&W G3 300 MHz with about 600 meg of RAM, so the machines are comparable (both pathetic by modern standards, but comparable). So, it actually appears that the Mac is slow at browsing, and IE works in such a way to emphasize that slowness, making it seem unbearably slow.
Also, a lot of apps, and Finder, aren't as threaded as they could be. While IE, for instance, is busy getting that big slashdot page ready to display, the dreaded spinning color-ball shows up, so you can't switch back and view the other pages you were reading.
Finally, much Apple software IS slow. There's a thread on comp.sys.mac.advocacy about this right now, where someone was saying that the new generation of iApps seem slower than the previous iApps, and pointing out an apparent correlation between those written in Carbon (fast) and Cocoa (slow). However, other people have pointed out examples of fast Cocoa apps, so that is not the problem. Most interesting was someone who wrote their own photo manager, and compared to iPhoto. For some things, his is 2 orders of magnitude faster than iPhoto. Evidently, Apple simply used crappy algorithms in iPhoto. Apple's mail program is similarly problematic when mailboxes get large. A lot of people on comp.sys.mac.advocacy have given up on it and switched to Eudora, and report their Macs are nice and fast at mail then.