There still is hope that GnuPG will be turned into something that can catch on with the masses (just like there's hope, however faint, that things like GNOME and KDE will catch on with the masses).
Is there any way that GnuPG could be built with a nice GUI for Windows? The fact is that for the time being, encryption will be worthless if the Windows users can't get the software.
Magazines (for the content based argument) seem to sell for approximately what they always have. What else out there continues to climb in price year after year?
Magazines have historically tended to increase in price when the advertising market is soft. One portion of the price of magazines is the cover/subscription price. The rest is the ad pages. Magazines generally don't like raising subscription/cover prices (as circulation declines), so if their costs increase dramatically, they look to sell more ad pages. But when they're having trouble selling ad pages to begin with, the only solution becomes to increase the prices paid by readers (although some "insider discounts" may be the first to go). The reason why magazines have seemed constant over the last 5 years is the strength of the ad market, largely led by the flood of dot-com money (think for a moment about how much your average high-profile startup burned with magazine and TV ads).
Maybe subscribers should be able to see a list of advertisers, with a checkbox allowing them to change the policy from default to explicit allow or deny. The frequency with which allowed ads show up could be based on the relative chances of an ad from that advertiser showing up on an unsubscribed user's page (so if ThinkGeek appears on 10% of the pages of a non-subscribed user, then every tenth page-view where an ad would have been supressed has a ThinkGeek ad). Slashdot then gets the best of both worlds, and it's almost a moderation system for ads.
The nearly undebatable fact is that people aren't going to be willing to subscribe in the numbers required to keep Slashdot afloat if the current form. My proposal was quite simple: give the paying subscribers earlier access to articles and comments. This in no way affect anyone's ability to post. Remember, moderators will probably be subscribing less than average, as the subscribers may tend towards the obsessive-compulsive reloaders, who are filtered out of the moderation pool. So the majority of moderators will see the article with the non-subscribers. So the subscribers get to post to an article for 5 minutes, during which time no one except subscribers can read it. I don't see the problem with that.
As for waiving the karma cap and allowing higher starting scores, I also see no problem. First of all, they've earned the karma. It's not like the ability of trolls to post at +3 is on sale.
What you're forgetting is that ThinkGeek does not really help Slashdot. Their both owned by OSDN. No money is paid. The ThinkGeek ad is, in a major sense, filler for ad-space they couldn't sell.
Part of the reason why Slashdot's ad sales suck is the pricing. Looking at the OSDN rate card, I see that the average CPM of the different plans is $65.81. Jesus Christ, that's steep. As a comparison, Google's ads are most definitely less than $30 CPM (and are possibly better targeted... the $30 figure is out of my ass, but when they debuted AdWords a few years ago, the top rate was $18. Since then ad-prices have generally fallen). Now, I'll ask you this: is Slashdot twice as effective at reaching a techie audience as Google?
What percentage of OSDN advertising space is currently unfilled? 90%? If it's 90% unfilled that means that reducing the price might cause it to be more filled. If OSDN dropped their rates to $12 (a five-fold drop in price), would they get five times the ads? I'd tend to think so.
The idiocy of the people at Andover/VA is truly stunning. They're still convinced that it's 1999, and if they set the ad rates stratospherically high, they'll sell them and make meeeeeeellions of dollars! I, a freshman CS major at UMass, could do a better job than those morons.
Basically the idea is to mod_rewrite (an Apache module that allows url manipulations) to redirect pages that commonly are requested by zombied Windows systems (/MSADC,/cmd.exe, etc) to a suitable site, mainly www.microsoft.com. I'm not sure whether the zombied machines will obey the 302, but as a symbolic gesture, it got some annoyance at being attacked by Code Red out of the way.
Last up, I'm gonna talk a little about advertisements and subscriptions. Slashdot continues to grow: our traffic has increased by like 10% in the last few months, and simply selling the banner ads you see on top of each page isn't going to be enough to keep us afloat if we keep growing. And selling banner ads in 2001 is an awful lot harder then it was in 1999.
The change will be a different ad size on the article page. Currently we have the standard banner size on top of all pages, but soon the article pages will instead have those huge square things that you see on CNet or ZD. I know this will be unpopular with many people, myself included, but when we make the switch, we will also have some sort of subscription system where you can pay a fee to disable them honestly. (No I don't know how much yet!)
Just to shut down the conspiracy theorists, nobody is forcing us to make these changes: The navbar. The new ad formats. The subscription system. I could just say 'No' to changes like these. But Slashdot is now four years old... and I want it to still be here four years from now. I hope you can understand the expensive reality associated with making this site happen every day for a quarter of a million readers.
One change which could be made (and would it be noticed? It could be in place now, for all I know) would be to have paying subscribers get a "live" feed (aka as soon as it's posted, subscribers can read and reply). Logged-in users get a 5 minute delay. AC's get a 15 minute delay. This eliminates some of the first post conditions (maybe a prohibition of anonymous posting in the first 5 minutes or until, say 10 logged comments have been posted is a good idea...), and means that if the trolls want to post early, they'll probably have to pay.
If SlashCode is so good, whats wrong with just using SlashDot as a forum for discussing things? Not up to it?
No web-board system could match IRC in the real-time department.
That said, an ongoing meta discussion would be the best, with occasional real-time IRC. Maybe the meta discussion could be limited to paying subscribers.
I myself think that the idea of paying for ad-free views is a bad idea. The smarter thing is to charge subscriptions for extra features to the site, with ads not being one of them. Some features that might be interesting could be: ability of paying subscribers with 75 karma to post at 3 (and be modded down accordingly); voting on rejected submissions; more regular metamod. Maybe paying subscribers who reach really high karma can become editors...
Yes I know that these will make Slashdot into a class system. But at least people will have an opportunity to join the ruling class; as it stands now, it's a two-class system, with a gulf between the editors and the readers/posters. Bridging this gap is a good thing, imho.
Hmm. You shake hands with your right hand. And you wear your watch on your left hand. Thus shaking hands will not bring your watch any closer to the other guy's watch than simply standing behind him in a queue. So where's the problem?
Sure, that's part of it, but Mac OS X is pretty tight, out of the box. No services enabled until you turn them on, and the/Applications,/bin,/usr/bin, and similar directories are all root/ admin, -rwxr-x-r-x by default.
I assumed that. That's why I put in the bit on "especially not running Mac OS X)"...
It's not so much that MyISAM is unreliable, as it doesn't work that well for a table that is being updated constantly, what with it's primitive locking setup. InnoDB supports transactions and consistent reads throughout a transaction.
In hindsight, MyISAM probably has more to do with Slashdot's tendency to slow down massively (because an update to a table essentially blocks all reads from that table.
Opera has had tabbed browsing since the early days (3.0? earlier?). I believe it's possible to edit the search feature so that web shortcuts are supported.
I'm sure I'm forgetting a bunch. My only gripe is that Opera sometimes crashes, although the newest version 6.0 B1 hasn't crashed on me once yet (although it has only been released a few days ago).
Opera crashes every few hours of heavy use. I think it might be in the ram caching code...
Assuming Moore's law holds for RAM (which I believe it does...), 640k 20 years ago was the equivalent of ~5GB. How many Slashdot readers have 5GB on a workstation?
ew worries about viruses, [because] the likelihood of getting a real virus
on the Mac is pretty low
Is it just me or is this (especially on Macs that aren't running OS X) more a function of the fact that few Mac viruses are written because the potential damage is so minimal relative to a Windows virus?
It appears that these three pictures are all green-lit, and are in progress at one stage or another.
Exactly. It takes at least three years to make an animated movie (from initial concept to release). Disney wants a Pixar movie every year, because it's the only part of Disney's animation business that does well. I mean, the last few non-Pixar Disney movies have been total flops.
I posted that while logged in... If Taco wants me to pay for this site, the login code should be fixed. If I'm paying for a login, I want to be logged in. I don't know it its a case of giving the wrong cookies, or the DB backend is being forgetful (that's what you get for running your business on MyISAM tables... if you're going to use MySQL, at least use InnoDB...), or what, but this needs to be fixed.
Detrermining which host if the fastest so the majority of the file is downloaded from them would be a bit more tricky though. You'd have to have a test download file, or calculate who is the fastest after the file has started downloading. Either way it is all possible, and really wouldn't be that hard.
How difficult would it be to add a header for Gnutella that allows start and end offsets to be specified? If you did it this way, you could do simultaneous transfers as follows:
To start, all n sources are treated equally and requests for (fize-size/(n*k)) bytes are sent to each source, with k being a subdividing coefficient (each request is actually for a small portion of the workload allocated. I will call each portion a "bundle")
As soon as one source's workload is completed, all bundles that have yet to be requested are divided evenly amongst the remaining.
Repeat the process.
This setup has the advantage of being able to dynamically adapt to network changes. If a cable gets ripped out of the wall, it can fall over dynamically. To add a new source for this file, you can just reallocate the unrequested bundles.
I seem to remember that MS's approach has tended to be, when faced with an either-or decision regarding future tech directions, to have one group push one side and another group push the other side. Thus:
32-bit PC OS: Windows NT vs. OS/2
Multimedia: Internet vs. CD-ROM
GUIs: Windows vs. Macintosh
This is a smart practice, as it provides a hedge. If the Mac really takes off again, Microsoft is there and will make lots of money off of Office for OS X. I myself have begun to think that what Microsoft is planning to do is to buy a Linux distributor and rebrand it as MS Linux. They could roll their own non-free toolkit (say, port MFC over) and write Office, IE, etc. for that toolkit. In theory, they could even build a distribution where the only GPL bit was the kernel.
Slashdot Mirror
Is there any way that GnuPG could be built with a nice GUI for Windows? The fact is that for the time being, encryption will be worthless if the Windows users can't get the software.
Magazines have historically tended to increase in price when the advertising market is soft. One portion of the price of magazines is the cover/subscription price. The rest is the ad pages. Magazines generally don't like raising subscription/cover prices (as circulation declines), so if their costs increase dramatically, they look to sell more ad pages. But when they're having trouble selling ad pages to begin with, the only solution becomes to increase the prices paid by readers (although some "insider discounts" may be the first to go). The reason why magazines have seemed constant over the last 5 years is the strength of the ad market, largely led by the flood of dot-com money (think for a moment about how much your average high-profile startup burned with magazine and TV ads).
Maybe subscribers should be able to see a list of advertisers, with a checkbox allowing them to change the policy from default to explicit allow or deny. The frequency with which allowed ads show up could be based on the relative chances of an ad from that advertiser showing up on an unsubscribed user's page (so if ThinkGeek appears on 10% of the pages of a non-subscribed user, then every tenth page-view where an ad would have been supressed has a ThinkGeek ad). Slashdot then gets the best of both worlds, and it's almost a moderation system for ads.
The nearly undebatable fact is that people aren't going to be willing to subscribe in the numbers required to keep Slashdot afloat if the current form. My proposal was quite simple: give the paying subscribers earlier access to articles and comments. This in no way affect anyone's ability to post. Remember, moderators will probably be subscribing less than average, as the subscribers may tend towards the obsessive-compulsive reloaders, who are filtered out of the moderation pool. So the majority of moderators will see the article with the non-subscribers. So the subscribers get to post to an article for 5 minutes, during which time no one except subscribers can read it. I don't see the problem with that.
As for waiving the karma cap and allowing higher starting scores, I also see no problem. First of all, they've earned the karma. It's not like the ability of trolls to post at +3 is on sale.
What you're forgetting is that ThinkGeek does not really help Slashdot. Their both owned by OSDN. No money is paid. The ThinkGeek ad is, in a major sense, filler for ad-space they couldn't sell.
Part of the reason why Slashdot's ad sales suck is the pricing. Looking at the OSDN rate card, I see that the average CPM of the different plans is $65.81. Jesus Christ, that's steep. As a comparison, Google's ads are most definitely less than $30 CPM (and are possibly better targeted... the $30 figure is out of my ass, but when they debuted AdWords a few years ago, the top rate was $18. Since then ad-prices have generally fallen). Now, I'll ask you this: is Slashdot twice as effective at reaching a techie audience as Google?
What percentage of OSDN advertising space is currently unfilled? 90%? If it's 90% unfilled that means that reducing the price might cause it to be more filled. If OSDN dropped their rates to $12 (a five-fold drop in price), would they get five times the ads? I'd tend to think so.
The idiocy of the people at Andover/VA is truly stunning. They're still convinced that it's 1999, and if they set the ad rates stratospherically high, they'll sell them and make meeeeeeellions of dollars! I, a freshman CS major at UMass, could do a better job than those morons.
Basically the idea is to mod_rewrite (an Apache module that allows url manipulations) to redirect pages that commonly are requested by zombied Windows systems (/MSADC, /cmd.exe, etc) to a suitable site, mainly www.microsoft.com. I'm not sure whether the zombied machines will obey the 302, but as a symbolic gesture, it got some annoyance at being attacked by Code Red out of the way.
You're quite right... Should have looked at the docs... It was sending out 302's though...
Yeah, I know this is offtopic...
From this article on October 22, 2001:
One change which could be made (and would it be noticed? It could be in place now, for all I know) would be to have paying subscribers get a "live" feed (aka as soon as it's posted, subscribers can read and reply). Logged-in users get a 5 minute delay. AC's get a 15 minute delay. This eliminates some of the first post conditions (maybe a prohibition of anonymous posting in the first 5 minutes or until, say 10 logged comments have been posted is a good idea...), and means that if the trolls want to post early, they'll probably have to pay.
No web-board system could match IRC in the real-time department.
That said, an ongoing meta discussion would be the best, with occasional real-time IRC. Maybe the meta discussion could be limited to paying subscribers.
I myself think that the idea of paying for ad-free views is a bad idea. The smarter thing is to charge subscriptions for extra features to the site, with ads not being one of them. Some features that might be interesting could be: ability of paying subscribers with 75 karma to post at 3 (and be modded down accordingly); voting on rejected submissions; more regular metamod. Maybe paying subscribers who reach really high karma can become editors...
Yes I know that these will make Slashdot into a class system. But at least people will have an opportunity to join the ruling class; as it stands now, it's a two-class system, with a gulf between the editors and the readers/posters. Bridging this gap is a good thing, imho.
Unless you're a Freemason! ;o)
A quasi-prominent Slashdotter, Eric Krout, has launched a Slash-like site that has a greater focus on Linux and free software news: MonoLinux
I assumed that. That's why I put in the bit on "especially not running Mac OS X)"...
It's not so much that MyISAM is unreliable, as it doesn't work that well for a table that is being updated constantly, what with it's primitive locking setup. InnoDB supports transactions and consistent reads throughout a transaction.
In hindsight, MyISAM probably has more to do with Slashdot's tendency to slow down massively (because an update to a table essentially blocks all reads from that table.
Opera has had tabbed browsing since the early days (3.0? earlier?). I believe it's possible to edit the search feature so that web shortcuts are supported.
My thoughts exactly. It's definitely more useful to the average Slashdot reader than a Linux 2.5 release (stable releases are a different story.
Opera crashes every few hours of heavy use. I think it might be in the ram caching code...
Assuming Moore's law holds for RAM (which I believe it does...), 640k 20 years ago was the equivalent of ~5GB. How many Slashdot readers have 5GB on a workstation?
How are we going to get a pad out in space that far?
*rimshot*
Is it just me or is this (especially on Macs that aren't running OS X) more a function of the fact that few Mac viruses are written because the potential damage is so minimal relative to a Windows virus?
Exactly. It takes at least three years to make an animated movie (from initial concept to release). Disney wants a Pixar movie every year, because it's the only part of Disney's animation business that does well. I mean, the last few non-Pixar Disney movies have been total flops.
I posted that while logged in... If Taco wants me to pay for this site, the login code should be fixed. If I'm paying for a login, I want to be logged in. I don't know it its a case of giving the wrong cookies, or the DB backend is being forgetful (that's what you get for running your business on MyISAM tables... if you're going to use MySQL, at least use InnoDB...), or what, but this needs to be fixed.
Just a spelling correction: s/fize/file/
How difficult would it be to add a header for Gnutella that allows start and end offsets to be specified? If you did it this way, you could do simultaneous transfers as follows:
- To start, all n sources are treated equally and requests for (fize-size/(n*k)) bytes are sent to each source, with k being a subdividing coefficient (each request is actually for a small portion of the workload allocated. I will call each portion a "bundle")
- As soon as one source's workload is completed, all bundles that have yet to be requested are divided evenly amongst the remaining.
- Repeat the process.
This setup has the advantage of being able to dynamically adapt to network changes. If a cable gets ripped out of the wall, it can fall over dynamically. To add a new source for this file, you can just reallocate the unrequested bundles.I seem to remember that MS's approach has tended to be, when faced with an either-or decision regarding future tech directions, to have one group push one side and another group push the other side. Thus:
- 32-bit PC OS: Windows NT vs. OS/2
- Multimedia: Internet vs. CD-ROM
- GUIs: Windows vs. Macintosh
This is a smart practice, as it provides a hedge. If the Mac really takes off again, Microsoft is there and will make lots of money off of Office for OS X. I myself have begun to think that what Microsoft is planning to do is to buy a Linux distributor and rebrand it as MS Linux. They could roll their own non-free toolkit (say, port MFC over) and write Office, IE, etc. for that toolkit. In theory, they could even build a distribution where the only GPL bit was the kernel.