Dillo and Firebird aren't really in the same class.
Dillo has a far faster layout engine (thanks a small amount to moi:-) ), uses much less RAM...but also has far fewer features. It would be a good browser for a PDA.
Anyone used Privoxy and sleezeball (runs off of squid)? I used to use squid/sleezeball, but for a single workstatop, Mozilla's/Firebird's built-in blocking just has less overhead.
I just don't see any behemoth coming out of the woodwork to stop MS dominance in this arena.
IBM? The OS/2 soul still burns -- they'd dearly love to reclaim their dominance -- and they can wait as long as is necessary for Linux to get nice and attractive on the desktop.
The problem with Microsoft is that because they're a monpolist (well, and because Slashdot doesn't like 'em, frequently for good reason), *any* deviation from published standards gets 'em raked through the coals. I doubt Mozilla, Opera, Konq, etc are fully standards-compliant either. Linux certainly isn't -- Linux says "this POSIX standard is broken", and it just gets ignored. The thing is, they don't catch flak for it.
So while I agree that "embrace and extend" *is* a real tactic that Microsoft has used historically, every time they deviate from a standard, they aren't deliberately out to get folks.
In good news for Mozilla, once a Microsoft product starts to stagnate, it tends to stay stagnant. So if the Moz people can keep trudging along, AOL or Dell or someone can ship Windows bundled with Mozilla (or Linux just plain catches on on the desktop), they may have a much better shot.
Microsoft dissolves development teams once a development project is over, and can have a tough time finding people to start up a long-dormant project. The Samba people have said it before in frustration, when they tried tracking down a Microsoft SMB developer to answer a question at a networking conference. There just wasn't anyone left who *knew* how Microsoft's SMB implementation worked. The Samba lead said in frusteration something along the lines that they knew Microsoft's SMB implementation better than anyone left at Microsoft.
In other news, water runs downhill, the sun rises in the east, and Dubya is dumb as a brick.
Actually, I haven't ever really seen much "innovation" from Microsoft, as Linus pointed out in his recen tinterview. Microsoft is not a particularly innovative company. They're a good publisher and good and monopoly management, but most of their products were purchased from someone else *after* they were developed and did well. (Folks could learn something from this -- the way to succeed in business just isn't usually small innovative engineering firms, but companies that let other companies try things out, make mistakes, and then just purchase the ones doing well (yes, at a more expensive price, but sans all the deadweight of failing companies).
I'm wondering if anyone either has real-world experience or benchmarks on optimium filesystems/partitions for a Linux workstation (the usual: web browsing, games, multimedia, etc)...
Just about all of the filesystems are excellent. Frankly, the degree of tweaking people do to try to maximize performance is ridiculous. These are all good filesystems, and none of them drastically better than the others when it comes to performance. This is *especially* true when you're just using the thing on a plain ol' vanilla workstation, where the load you put on your filesystems is essentially nonexistant.
I'd make my decision based on features.
I suspect most folks want journalling support. It's an awfully nice feature to have. So probably ext2 isn't a great choice for them. You can easily migrate an ext2 filesystem to ext3, so if you're currently using ext2, ext3 is definitely the way to go. Reiser can deal particularly well with non-empty-but-extremely-small-files, and drastically reduces space usage for such files. Source code doesn't take as much space (percentagewise) of a drive as it used to, but it might be a valuable consideration for developers). XFS had a fairly specific design goal (and here I will delve into performance, since it has a particular specialization) -- streaming files linearly.
I'm not sure what unique features JFS has, since I haven't played with it.
That affects *my* software, which is only a fraction of a percent of the GPL software out there. Many engineers have simply blindly put their faith in the GPL.
Keep in mind that Stallman (well, the FSF) still has ahold of the key to the most incredibly vast storehouse of IP in the world -- control of the GPL. In a worst-case scenerio (and it would take a seriously doomsday scenerio), Stallman would probably just make GPLv3 have a clause stating that SCO and SCO alone may not use GPLed code.
This is the one thing that terribly worries me. The FSF has massive and growing IP control for perpetuity. Now, I happen to think that as long as Stallman doesn't go senile or have a revelation or something, GPL revisions won't be nasty. But what if an FSF member gets paid off to release a "revision" of the GPL that BSD-licenses all GPLed code? Sure, Stallman would argue that such a modification is illegitimate, but who's to say what a judge would rule as being legal? My money would be with Stallman, but it's a thought. If I were Microsoft, faced with losing a monopoly, I might want to try such a gamble.
And what if someone tried buying out members of the FSF with a couple million each for a "legitimate" GPL to BSD conversion? Control of all GPLed software would be a phenomenally valuable asset. Maybe RMS wouldn't be bought out, but he won't be alive forever, and the GPL does nothing but spread. What will happen 50 years down the line?
It's probably vast overkill, but 3d systems frequently don't lend themselves to 2d visualization that well. Take, say, a building's power system. Sure, you could pop it on someone's computer with a 3d card, but computer input devices aren't anywhere near as nice to use as actually moving over to what you want to look at.
It might be interesting in computer vision -- you can actually produce an environment to desired specifications (or over a gradient of worsening conditions) to test in.
And email is not as reliable as other services. It's not regular mail, there's no certified letters.
I had an airmail letter get lost just the other day.
Considering the far, far, far, far larger volume of email I send -- way over 100 to 1 ratio to normal mail -- email may not actually be less reliable.
Re:IRC Needs Improvements-but you missed security
on
IRC in the Dog House?
·
· Score: 1
Ok... Security 101 class - what does a public key give you on IRC that a nick doesn't ???
You also sign a timestamped message from the server when you sign in. By checking signatures and comparing the advertised private key, other clients can form trusted mappings between "(bob37, [pubkey])" pair on Wednesday and a "(bob37, [pubkey])" pair today. This means that I know that Bob really is bob.
Private keys wouldn't need to be broadcast on a per message basis. Once a trusted mapping has been formed, the clients know that "bob37" is trusted to be "Mike Smith, CEO of SmithCorp" for the length of this session.
What are you protecting from when you encrypt between you and the server (lets assume you also ment authenticate as well - since encryption without authentication is worthless)
Those on the same network and those at your ISP monitoring everything you say and where you're talking to. When paired with an anonymity system, you and what you say cannot be easily mapped to you, regardless of whether a privacy attacker comes from your ISP or the IRC network. For an untrusted IRC server to be dealt with, addtional direct connections (not passing through the IRC server) would need to be established -- hence the "privacy ring" I mentioned earlier.
Verisign made a business move. We all know and agree that there are technically superior ways to provide the same service -- from within the browser.
At the very least, Verisign has only moved to a local maxima -- they *may* have improved the overall user experience somewhat, but had they *not* done so, more significant user experience improvements could have been provided.
Frankly, I think it's more that consumers don't realize the magnitude and impact to them of what Verisign is doing, so they aren't complaining because they only occasionally see the drawbacks.
Weeeellll...while I agree entirely with most of your points, I'm not sure I agree that
Smooshing the layers together, for commercial benefit or otherwise, is not innovation.
There is significant benefit to several types of networks in not having clear separation of layers. Most ad-hoc networking systems maintain some type of routing cache -- such a cache would exist and operate at the network layer, but would probably benefit from utilizing information only present in the transport layer (you probably wouldn't want to throw out cached routing information for an open TCP connection).
A similar argument goes for IPSec and similar network-level security protocols.
In wireless protocols, it may be beneficial to utilize information from the transport layer (don't bother retransmitting UDP packets, for instance).
The isolated-layer structure provides for simpler state machines. This is a big deal -- they're easier to write, and easier to avoid bugs in. However, sometimes useful information is thrown out in such a situation.
I'm dubious. I read one of the very first, if not the first, announcements on NANOG, and it was after-the-fact (or approximately at-the-fact). There was certainly not the at *least* six months of warning that would be necessary for a change like this.
Also, has anyone noticed that Verisign and Microsoft, two of the largest tech companies that play the nastiest and are the *worst* at coming out with new tech (actually, to be honest, I don't know Verisign's history well enough to know whether they're historically like this or not) are the ones that *constantly* claim that any interference with their operations would "stifle innovation"?
I don't see Google pulling the "stifle innovation" card. Google's research lab comes out with exciting, helpful, interesting things on a regular basis. I don't see Apple pulling the "stifle innovation" card -- and while Apple may not be the tech luminary that it once was, it still comes out with decent and out-of-the-ordinary products. The only people claiming that their "innovation" is at risk are those who *aren't* innovating.
Slashdot Mirror
Intuit has no monopoly.
Microsoft does.
It doesn't take all that much pondering.
Dillo and Firebird aren't really in the same class.
:-) ), uses much less RAM...but also has far fewer features. It would be a good browser for a PDA.
Dillo has a far faster layout engine (thanks a small amount to moi
(And by the both Privoxy and sleezeball comment, I meant that I'm curious as to how they compare, in folks' opinion.)
XUL is still flaky, still misses mouse clicks, dumps menus in the wrong spot, etc.
I'm *using* Firebird at the moment, too.
This would be just desserts for MS, which forced Opera in to pretending it was MSIE by default, so that hotmail and MSN would work...
Err...MS used to pretend that IE was Navigator. Nobody's innocent here.
Anyone used Privoxy and sleezeball (runs off of squid)? I used to use squid/sleezeball, but for a single workstatop, Mozilla's/Firebird's built-in blocking just has less overhead.
I just don't see any behemoth coming out of the woodwork to stop MS dominance in this arena.
IBM? The OS/2 soul still burns -- they'd dearly love to reclaim their dominance -- and they can wait as long as is necessary for Linux to get nice and attractive on the desktop.
Work is not about pleasing the customer. It's about pleasing one's boss. A boss of a web designer is going to get crabby if an element is 1mm off.
The problem with Microsoft is that because they're a monpolist (well, and because Slashdot doesn't like 'em, frequently for good reason), *any* deviation from published standards gets 'em raked through the coals. I doubt Mozilla, Opera, Konq, etc are fully standards-compliant either. Linux certainly isn't -- Linux says "this POSIX standard is broken", and it just gets ignored. The thing is, they don't catch flak for it.
So while I agree that "embrace and extend" *is* a real tactic that Microsoft has used historically, every time they deviate from a standard, they aren't deliberately out to get folks.
In good news for Mozilla, once a Microsoft product starts to stagnate, it tends to stay stagnant. So if the Moz people can keep trudging along, AOL or Dell or someone can ship Windows bundled with Mozilla (or Linux just plain catches on on the desktop), they may have a much better shot.
Microsoft dissolves development teams once a development project is over, and can have a tough time finding people to start up a long-dormant project. The Samba people have said it before in frustration, when they tried tracking down a Microsoft SMB developer to answer a question at a networking conference. There just wasn't anyone left who *knew* how Microsoft's SMB implementation worked. The Samba lead said in frusteration something along the lines that they knew Microsoft's SMB implementation better than anyone left at Microsoft.
In other news, water runs downhill, the sun rises in the east, and Dubya is dumb as a brick.
Actually, I haven't ever really seen much "innovation" from Microsoft, as Linus pointed out in his recen tinterview. Microsoft is not a particularly innovative company. They're a good publisher and good and monopoly management, but most of their products were purchased from someone else *after* they were developed and did well. (Folks could learn something from this -- the way to succeed in business just isn't usually small innovative engineering firms, but companies that let other companies try things out, make mistakes, and then just purchase the ones doing well (yes, at a more expensive price, but sans all the deadweight of failing companies).
I'm wondering if anyone either has real-world experience or benchmarks on optimium filesystems/partitions for a Linux workstation (the usual: web browsing, games, multimedia, etc)...
Just about all of the filesystems are excellent. Frankly, the degree of tweaking people do to try to maximize performance is ridiculous. These are all good filesystems, and none of them drastically better than the others when it comes to performance. This is *especially* true when you're just using the thing on a plain ol' vanilla workstation, where the load you put on your filesystems is essentially nonexistant.
I'd make my decision based on features.
I suspect most folks want journalling support. It's an awfully nice feature to have. So probably ext2 isn't a great choice for them. You can easily migrate an ext2 filesystem to ext3, so if you're currently using ext2, ext3 is definitely the way to go. Reiser can deal particularly well with non-empty-but-extremely-small-files, and drastically reduces space usage for such files. Source code doesn't take as much space (percentagewise) of a drive as it used to, but it might be a valuable consideration for developers). XFS had a fairly specific design goal (and here I will delve into performance, since it has a particular specialization) -- streaming files linearly.
I'm not sure what unique features JFS has, since I haven't played with it.
Then they just grep based on host referrer.
That affects *my* software, which is only a fraction of a percent of the GPL software out there. Many engineers have simply blindly put their faith in the GPL.
I've never owned a Sony console, but my understanding is that "PSX" is the conventional acronym (for *some* reason) for the Playstation 1.
Why would Sony produce so much confusion by calling this new system (an enhanced PS2, apparently) the same thing?
Impossible. Many may change their mind, be uncontactable, be dead, etc. Do you have the geneaology of every line of code out there? I certainly don't.
Keep in mind that Stallman (well, the FSF) still has ahold of the key to the most incredibly vast storehouse of IP in the world -- control of the GPL. In a worst-case scenerio (and it would take a seriously doomsday scenerio), Stallman would probably just make GPLv3 have a clause stating that SCO and SCO alone may not use GPLed code.
This is the one thing that terribly worries me. The FSF has massive and growing IP control for perpetuity. Now, I happen to think that as long as Stallman doesn't go senile or have a revelation or something, GPL revisions won't be nasty. But what if an FSF member gets paid off to release a "revision" of the GPL that BSD-licenses all GPLed code? Sure, Stallman would argue that such a modification is illegitimate, but who's to say what a judge would rule as being legal? My money would be with Stallman, but it's a thought. If I were Microsoft, faced with losing a monopoly, I might want to try such a gamble.
And what if someone tried buying out members of the FSF with a couple million each for a "legitimate" GPL to BSD conversion? Control of all GPLed software would be a phenomenally valuable asset. Maybe RMS wouldn't be bought out, but he won't be alive forever, and the GPL does nothing but spread. What will happen 50 years down the line?
It's probably vast overkill, but 3d systems frequently don't lend themselves to 2d visualization that well. Take, say, a building's power system. Sure, you could pop it on someone's computer with a 3d card, but computer input devices aren't anywhere near as nice to use as actually moving over to what you want to look at.
It might be interesting in computer vision -- you can actually produce an environment to desired specifications (or over a gradient of worsening conditions) to test in.
Commercial 3d virtual tourism?
I believe Outlook's scheme is standardized, and that other clients support it.
And email is not as reliable as other services. It's not regular mail, there's no certified letters.
I had an airmail letter get lost just the other day.
Considering the far, far, far, far larger volume of email I send -- way over 100 to 1 ratio to normal mail -- email may not actually be less reliable.
Ok... Security 101 class - what does a public key give you on IRC that a nick doesn't ???
You also sign a timestamped message from the server when you sign in. By checking signatures and comparing the advertised private key, other clients can form trusted mappings between "(bob37, [pubkey])" pair on Wednesday and a "(bob37, [pubkey])" pair today. This means that I know that Bob really is bob.
Private keys wouldn't need to be broadcast on a per message basis. Once a trusted mapping has been formed, the clients know that "bob37" is trusted to be "Mike Smith, CEO of SmithCorp" for the length of this session.
What are you protecting from when you encrypt between you and the server (lets assume you also ment authenticate as well - since encryption without authentication is worthless)
Those on the same network and those at your ISP monitoring everything you say and where you're talking to. When paired with an anonymity system, you and what you say cannot be easily mapped to you, regardless of whether a privacy attacker comes from your ISP or the IRC network. For an untrusted IRC server to be dealt with, addtional direct connections (not passing through the IRC server) would need to be established -- hence the "privacy ring" I mentioned earlier.
Or a decision on whether the short-term benefit of the toy is worth the medium term benefit of sleeping on the couch.
Okay, let me put it this way.
Verisign made a business move. We all know and agree that there are technically superior ways to provide the same service -- from within the browser.
At the very least, Verisign has only moved to a local maxima -- they *may* have improved the overall user experience somewhat, but had they *not* done so, more significant user experience improvements could have been provided.
Frankly, I think it's more that consumers don't realize the magnitude and impact to them of what Verisign is doing, so they aren't complaining because they only occasionally see the drawbacks.
Weeeellll...while I agree entirely with most of your points, I'm not sure I agree that
Smooshing the layers together, for commercial benefit or otherwise, is not innovation.
There is significant benefit to several types of networks in not having clear separation of layers. Most ad-hoc networking systems maintain some type of routing cache -- such a cache would exist and operate at the network layer, but would probably benefit from utilizing information only present in the transport layer (you probably wouldn't want to throw out cached routing information for an open TCP connection).
A similar argument goes for IPSec and similar network-level security protocols.
In wireless protocols, it may be beneficial to utilize information from the transport layer (don't bother retransmitting UDP packets, for instance).
The isolated-layer structure provides for simpler state machines. This is a big deal -- they're easier to write, and easier to avoid bugs in. However, sometimes useful information is thrown out in such a situation.
The proper way to implement this would be more along the lines of:
/* TODO: error-handling code */ }
:-) )
hostent_p = gethostbyname(addr);
if (!hostent_p) {
if (h_errno == HOST_NOT_FOUND) {
return USER_DEFINED_REDIRECT;
} else {
} else
return hostent_p->h_addr;
This code should only ever appear in a web browser (actually, it should only appear in a fork()-using web browser...
There WAS advance notice.
I'm dubious. I read one of the very first, if not the first, announcements on NANOG, and it was after-the-fact (or approximately at-the-fact). There was certainly not the at *least* six months of warning that would be necessary for a change like this.
Also, has anyone noticed that Verisign and Microsoft, two of the largest tech companies that play the nastiest and are the *worst* at coming out with new tech (actually, to be honest, I don't know Verisign's history well enough to know whether they're historically like this or not) are the ones that *constantly* claim that any interference with their operations would "stifle innovation"?
I don't see Google pulling the "stifle innovation" card. Google's research lab comes out with exciting, helpful, interesting things on a regular basis. I don't see Apple pulling the "stifle innovation" card -- and while Apple may not be the tech luminary that it once was, it still comes out with decent and out-of-the-ordinary products. The only people claiming that their "innovation" is at risk are those who *aren't* innovating.