Although software libre developers can't get access to Microsoft's specs or API source, it's quite possible to work around the problem.
Basically, a commercial entity or entities (e.g. Sun, IBM) with an interest in promoting free software competition to Microsoft could finance an operation which would use the Microsoft specs to put together a compliance suite. Since a great deal of the effort in reverse-engineering protocols lies in making sure that your tests are complete, a black-box compliance suite -- even though not itself free -- would dramatically ease the process of reverse-engineering the specs themselves.
As a dandy side-effect, the suite would also show up where Microsoft doesn't comply with their own specs, which detail might be of interest to the public and the Court.
Seriously though, what were they thinking? AOL _is_ a trademark.
Yes, it is. Therefore it's illegal for me to market my company "Another Old Look" (which sells faux antiques) as "AOL."
On the other hand, it does not prevent me from using their trademark to refer to them, as in the ISP help page, http://www.sample.com/help/former_aol_users which explains the Internet to people who have only used AOL.
Just as there is no trademark infringement if an auto parts manufacturer sells decorative wheel covers designed for Cadillacs and calls them "Roulette Wheel Covers for Cadillac," there is no trademark infringement for someone who provides accessories for AOL mentioning that fact. They're not claiming to be AOL, they're describing a context.
Hungarian Notation (with the embedding of C-language typing welded to all of the data in the system) is one of the main reasons that Microsoft had such an ugly time going from 16- to 32-bits and why MSWindows still isn't 64-bit clean.
Give samples and distribution rights to University LAN stations. A lot of University campuses have their own inside-the-firewall Net radio and are starved for material.
The big problem with POF has always been that it has higher loss and dispersion than glass. Until those are solved POF is still going to be limited to very short distances.
The journals accept contributions from academics, then have volunteer academics review them, publishing ths results. For this they charge the academics who provided the content and editorial screening a stiff price, as well as acquiring a 95-year lock against anyone else disseminating the research.
Yeah, I can see how that model is about as close to perfect as it could be.
then that variable 300-900 kbps is exactly what you'd see. The problem isn't between you and your ISP (which is what you'd be testing with line meters etc.) but with the ISP->backbone connection.
The real test, frankly, is to get bandwidth from someone with heavy-duty backbone connection (e.g. AT&T) and just plain hammer it with mondo file transfers scattered across the day. If your transfer times are varying with Net traffic periods, your ISP is the bottleneck.
You might be able to get similar information cheaper by doing repeated traceroutes and logging the delay between the ISP and the next router up, which would indicate the queueing delay at the ISP's routers.
This may be new in the USA
on
Shop Till It Drops
·
· Score: 5, Informative
but something very much like it is quite common in Japan. The last time I was there, there was a beast of a machine that sold everything from fast food to condoms in the lobby.
Selective enforcement of laws (2600, anyone?) allows them to selectively threaten people for leverage (e.g. making region-free players hard to get.)
Uniform enforcement, on the other hand, or even the widely-publicized appearance of uniform enforcement, brings the issues out of the geek ghetto to where the voting public confronts it.
Best thing that could happen would be for the RIAA to file criminal charges against Aunt Martha for letting her friends copy her Burl Ives recordings.
The government kinda shot itself in the foot with this one. It will be damn hard to prove that you have distribute works for 180 days whose total value is more than $999.99US.
Nyah, it's easy. The RIAA has someone download 2000 copies and they're there.
The problem isn't the hardware, it's the software drivers. In fact, the speed could be dramatically increased with revised software drivers. However, no manufacturer has presently made this aspect of driver performance a priority.
It looks like this is primarily a Microsoft-drivers problem. I wonder if anyone has looked at whether the XFree86 and DRM drivers could do better.
Unfortunately, I couldn't find any discussion of the issue on the OSI site, so I can't comment on any distinctive features of the proposed terms relative to, e.g., the GPL.
In the case of the most basic open source software, though, no license is really required. The author(s) retain copyright and simply choose to distribute the source along with (or even in place of) object code. It's only the perverse history of the last 25 years that leads us to think that an author gives up rights if he doesn't keep source code under lock and key.
Assume, then, that there are additional terms (again, see the GPL). The question is whether these additional terms are such that failure to execute a contract would vitiate them. In the case of the GPL, I don't see how that could be; perhaps someone could explain.
As for click-wrap, this implies some degree of license administration. All in all, it's probably much easier to simply keep a registration database of those who agree to the terms in return for redistribution or derivative-works rights.
PI lawyers won't waste time going after Aunt Martha; if any ambulances arrive late they're going for the deep pockets of the company which sold the boxen, runs the servers, wrote the software, and ignored seven years of warnings about exploits waiting to happen.
What's more, they're going to win. A jury will take nanoseconds to decide between a grieving family and a convicted abusive monopolist sitting on $40 billion in liquid assets.
As stated, it scans networks looking for infectable files. The process would go something like this:
1D10T introduces some kewl "screensaver" or whatever to a Win98 notebook while traveling
Goes home, lights up infected machine on corporate network
Machine spots some -rwxrwxrwx file on the network
Which is later executed by unsuspecting Linux user
For real fun, the mysadmin who let a world-writable executable exist invokes it while su root.
Alas, far too many *nix networks still have the implicit assumption that all of the machines connected to them are securely maintained. I know of at least one very large company where any machine on the network can get root NFS access by just spitting out the right packets -- and there are Win98 machines on the same network.
Friends who work at Philips (which bought Signetics years ago) tell me that the 25120 datasheet may be updated and reissued. Keep an eye out for it once Philips gets a usable website.
I recall an article in Analog back in the '60s that discussed the troubles that oil-eating bugs caused the military. It's not so much that they consume all of the fuel as that they denature it, cause acidic buildup leading to corrosion, and clog pumps and filters.
The US lost a fair number of aircraft to this kind of bacterial mischief back then before they learned to put antibacterials in the fuel. I'd hardly be surprised to find that the bioengineers have found ways to make bugs that like the antibacterials.
I love the productivity claims
on
.NETly News
·
· Score: 3, Funny
Wow! 50% improvement in programmer productivity.
Fine print:
... at shops like Microsoft where the entire design cycle consists of coding. In more mature shops where requirements analysis, specification, design, and QA take up 80-90% of the design cycle things may be a bit different.
Brin: When the government pursues new surveillance powers, our habit is to kick and scream and moan and then watch helplessly while they get what they want, as when something bad like 9/11 happens. A far more effective
technique is to demand fierce accountability measures in return for granting our servants the tools they claim they need. That?s how to keep both safety and freedom.
It's a shell game. In effect (I won't say it's deliberate) the focus on secrecy keeps the powerful in power. This is becausethe "watchdog" groups are so obsessed with secrecy that they ignore what the observers do with the information that they get regardless.
This works for both sides. The privacy lobby gets the ego boost and righteous publicity, while the watchers manage to get the information they want with minor restrictions on acquisition and very little constraint on its use.
You'll realize that the DMA's definition of "spam" is mass-mail from Somebody Else. About the only thing that the DMA policy requires of mainsleaze spammers is that they have "remove" addresses, and nobody trusts them anyway.
Bottom line: this is just another attempt to head off effective legislation by pretending "industry self-regulation."
This affects systems with telnet or rlogin accessible from the Internet? The implication is that these were somehow not vulnerable without this buffer overrun.
News to me.
We all know that the deal sucks, we all tend to agree that the deal-making process sucks, and most of us think that the deal-makers have serious personal problems.
Don't bother telling the Judge that part.
What we can tell her that the Court might actually listen to is this: how can Microsoft wiggle through loopholes? A consent decree is, when you get down to it, code. Legal rather than computer code, but code nonetheless. Let's apply the famous myriad eyeballs to finding bugs in the code here, and tell the Court in clear terms (as the Samba team have) just how it's broken.
Let's tell the Court what they don't know about this deal.
What happens around here is that the SOE consists of crippled P200 boxen with all storage over the LAN. Locked down to a fare-thee-well, of course, and Linux forbidden on pain of termination. No personal machines allowed to connect to the Company net (including personal boxen from home) and in "the present business conditions" no new Company boxen or notebooks.
So, of course, we all just use personal machines for our work (including personal notebook machines) and test to our little hearts' content at home or on our own personal notebooks.
Everybody wins: the workers get their work done without being hassled by IT and the Company saves a wad on all the machinery, licenses, etc. that it doesn't have to buy.
Backups are sometimes a hassle, and we haven't gotten the ad-hoc 802.11b network to do CVS reliably, but otherwise things are pretty decent.
Face it, the amount of heat transfer through the case itself is trivial. Stick a thermal probe on the the inside of the case, on the outside, and ambient. There's only a small fraction of a degree difference between the inside and outside of the steel, but quite a difference between the inside of the steel and the air leaving the case.
Al may be light and slick-looking, but it's not really a thermal improvement over steel. If you really want to improve heat transfer through the case, try these:
Paint the inside flat black. A remarkable amount of internal heat is transferred by radiation.
Put a thermally-conductive pad under the mobo so that more heat makes it to the steel directly.
Add another pad (or at least vent some air) between the mobo tray and the bottom panel (right side on towers.)
Slashdot Mirror
Basically, a commercial entity or entities (e.g. Sun, IBM) with an interest in promoting free software competition to Microsoft could finance an operation which would use the Microsoft specs to put together a compliance suite. Since a great deal of the effort in reverse-engineering protocols lies in making sure that your tests are complete, a black-box compliance suite -- even though not itself free -- would dramatically ease the process of reverse-engineering the specs themselves.
As a dandy side-effect, the suite would also show up where Microsoft doesn't comply with their own specs, which detail might be of interest to the public and the Court.
Yes, it is. Therefore it's illegal for me to market my company "Another Old Look" (which sells faux antiques) as "AOL."
On the other hand, it does not prevent me from using their trademark to refer to them, as in the ISP help page, http://www.sample.com/help/former_aol_users which explains the Internet to people who have only used AOL.
Just as there is no trademark infringement if an auto parts manufacturer sells decorative wheel covers designed for Cadillacs and calls them "Roulette Wheel Covers for Cadillac," there is no trademark infringement for someone who provides accessories for AOL mentioning that fact. They're not claiming to be AOL, they're describing a context.
Hungarian Notation (with the embedding of C-language typing welded to all of the data in the system) is one of the main reasons that Microsoft had such an ugly time going from 16- to 32-bits and why MSWindows still isn't 64-bit clean.
Give samples and distribution rights to University LAN stations. A lot of University campuses have their own inside-the-firewall Net radio and are starved for material.
The big problem with POF has always been that it has higher loss and dispersion than glass. Until those are solved POF is still going to be limited to very short distances.
Yeah, I can see how that model is about as close to perfect as it could be.
The real test, frankly, is to get bandwidth from someone with heavy-duty backbone connection (e.g. AT&T) and just plain hammer it with mondo file transfers scattered across the day. If your transfer times are varying with Net traffic periods, your ISP is the bottleneck.
You might be able to get similar information cheaper by doing repeated traceroutes and logging the delay between the ISP and the next router up, which would indicate the queueing delay at the ISP's routers.
but something very much like it is quite common in Japan. The last time I was there, there was a beast of a machine that sold everything from fast food to condoms in the lobby.
Uniform enforcement, on the other hand, or even the widely-publicized appearance of uniform enforcement, brings the issues out of the geek ghetto to where the voting public confronts it.
Best thing that could happen would be for the RIAA to file criminal charges against Aunt Martha for letting her friends copy her Burl Ives recordings.
Nyah, it's easy. The RIAA has someone download 2000 copies and they're there.
In the case of the most basic open source software, though, no license is really required. The author(s) retain copyright and simply choose to distribute the source along with (or even in place of) object code. It's only the perverse history of the last 25 years that leads us to think that an author gives up rights if he doesn't keep source code under lock and key.
Assume, then, that there are additional terms (again, see the GPL). The question is whether these additional terms are such that failure to execute a contract would vitiate them. In the case of the GPL, I don't see how that could be; perhaps someone could explain.
As for click-wrap, this implies some degree of license administration. All in all, it's probably much easier to simply keep a registration database of those who agree to the terms in return for redistribution or derivative-works rights.
What's more, they're going to win. A jury will take nanoseconds to decide between a grieving family and a convicted abusive monopolist sitting on $40 billion in liquid assets.
- 1D10T introduces some kewl "screensaver" or whatever to a Win98 notebook while traveling
- Goes home, lights up infected machine on corporate network
- Machine spots some -rwxrwxrwx file on the network
- Which is later executed by unsuspecting Linux user
- For real fun, the mysadmin who let a world-writable executable exist invokes it while su root.
Alas, far too many *nix networks still have the implicit assumption that all of the machines connected to them are securely maintained. I know of at least one very large company where any machine on the network can get root NFS access by just spitting out the right packets -- and there are Win98 machines on the same network.Friends who work at Philips (which bought Signetics years ago) tell me that the 25120 datasheet may be updated and reissued. Keep an eye out for it once Philips gets a usable website.
The US lost a fair number of aircraft to this kind of bacterial mischief back then before they learned to put antibacterials in the fuel. I'd hardly be surprised to find that the bioengineers have found ways to make bugs that like the antibacterials.
Wow! 50% improvement in programmer productivity.
Fine print:
... at shops like Microsoft where the entire design cycle consists of coding. In more mature shops where requirements analysis, specification, design, and QA take up 80-90% of the design cycle things may be a bit different.
Brin: When the government pursues new surveillance powers, our habit is to kick and scream and moan and then watch helplessly while they get what they want, as when something bad like 9/11 happens. A far more effective
technique is to demand fierce accountability measures in return for granting our servants the tools they claim they need. That?s how to keep both safety and freedom.
It's a shell game. In effect (I won't say it's deliberate) the focus on secrecy keeps the powerful in power. This is becausethe "watchdog" groups are so obsessed with secrecy that they ignore what the observers do with the information that they get regardless.
This works for both sides. The privacy lobby gets the ego boost and righteous publicity, while the watchers manage to get the information they want with minor restrictions on acquisition and very little constraint on its use.
You'll realize that the DMA's definition of "spam" is mass-mail from Somebody Else. About the only thing that the DMA policy requires of mainsleaze spammers is that they have "remove" addresses, and nobody trusts them anyway.
Bottom line: this is just another attempt to head off effective legislation by pretending "industry self-regulation."
This affects systems with telnet or rlogin accessible from the Internet? The implication is that these were somehow not vulnerable without this buffer overrun.
News to me.
The US took its first major step last month with the Microsoft settlement: MSN will be the "US Government online."
We all know that the deal sucks, we all tend to agree that the deal-making process sucks, and most of us think that the deal-makers have serious personal problems.
Don't bother telling the Judge that part.
What we can tell her that the Court might actually listen to is this: how can Microsoft wiggle through loopholes? A consent decree is, when you get down to it, code. Legal rather than computer code, but code nonetheless. Let's apply the famous myriad eyeballs to finding bugs in the code here, and tell the Court in clear terms (as the Samba team have) just how it's broken.
Let's tell the Court what they don't know about this deal.
What happens around here is that the SOE consists of crippled P200 boxen with all storage over the LAN. Locked down to a fare-thee-well, of course, and Linux forbidden on pain of termination. No personal machines allowed to connect to the Company net (including personal boxen from home) and in "the present business conditions" no new Company boxen or notebooks.
So, of course, we all just use personal machines for our work (including personal notebook machines) and test to our little hearts' content at home or on our own personal notebooks.
Everybody wins: the workers get their work done without being hassled by IT and the Company saves a wad on all the machinery, licenses, etc. that it doesn't have to buy.
Backups are sometimes a hassle, and we haven't gotten the ad-hoc 802.11b network to do CVS reliably, but otherwise things are pretty decent.
Al may be light and slick-looking, but it's not really a thermal improvement over steel. If you really want to improve heat transfer through the case, try these: