Kernel mode drivers need to be recompiled with each new release. Since the source of most drivers are already included in the kernel;s source tree this is not a problem.
Though, user-mode applications which everyone cares about have always been binary compatible with newer versions of the kernel. They only break if syscalls that the kernel exposes to user mode change, which they obviously have not. New syscalls are added with new releases of the kernel, but old ones are never removed or changed for compatibility reasons.
Screw the net... get them to search my freakin hdd. Its only 80gigs.
It's MUCH harder to keep track of a changing hard disk with 50 gazillion file formats, compared to what google does with the (relatively) slow changing net and indexing pdf, ps and html files.
Boa currently seems to be the favorite web server in the embedded crowd, and embedded Linux, despite all the marketing hype, really is a big deal. Supposedly, an older version of Boa, v0.92q, runs in 32K address space on m68k, like used in uCLinux
HAL the 15 month only computer brain
on
AI Going Nowhere?
·
· Score: 2, Informative
A Israel based research company is claiming to have created a computer a-i program that has passed the turning test for 15 month old baby: i.e when talking to it, you can not tell the difference between the computer and a 15 month old baby. By using a behaviorist approach (positive reinforemecent on desirable responses) they are planning to simulate and grow and adult mind in the next 10 years. The have video press releases here.
The slashdot heading is misleading; these guys are trying to make a system that will make it more expensive for mass emailers to send out spam. Not necessarily through money (although that is an option) but through exchange of resources, like cpu cycles. Very novel concept. I can imagine the same being done all over the internet. Instead of paying by cash you pay with something you have in abunadance (definately not cash)... cpu cycles.
Suppose slashdot wants to make revenue: every user that hits the slashdot page agrees to donate some of their cpu resources to slashdot. slashdot inturn sells these resource for cash to GENOME xyz company which uses distributed computing to fold DNA sequences.
The patch for MSSQL vulnerability was out 6 months before the hammer virus hit. so that too was fixed "relatively" quickly. as for half the internet not being slowed down by the vulnerability; maybe its because of this
> no need to bother with sequence number guessing, just send your data packet right away, and pretend the connection is already open.
wrong. you still need to know the sequence number of the stream from the client to the server if you are going to send a packet. otherwise the server will drop the packet.
>Too bad, though. Lack of competition will >decrease choice and progress. Intel can now slow >down it's R&D and it may be longer before we see >quantum chips and true AI. Worst of all, no >realistic simulated customized porno. There goes > my 3-breasted Klingon babe fantasy (sniff). > I'll miss you AMD.
Now Intel can take all the R&D they had in making x86 compatiable chips and move it to something other, like quantum chips. They dont need to ensure their domination and have more felxiblity to explore other alternatives than tired x86 line.
I dont know how hardcore you have to be to dish out 45 bucks for an entire year of subscription to the Xbox live service. i'm wondering if Microsoft is losing money on this as well...?
narrower flexible cables is all the SATA has to offer? its like switching from a two-headed screw driver standard to a philips one. the heads are different but the screw gose in the same way. give me a few features of SCSI in an affordable package and then your talking.
on the other hand: installing more ram/new-cpu wouldn't be such a pain.. 2 hard disks, a dvd and cd writer later your computer would have enough ribbons to host a toddlers birthday party. and to get rid of the freaking master/slave shite... ahhh. damn it I want SATA!
>Unfortunately, using Fat32 or NTFS(if you >were "Gasp" running nt/2k), you would be >repeatledy using the same flash sectors, >quickly burning them out.
Thats why M-System's Disk on chip has it's TrueFFS file system, which "Prolongs Flash Life Span: TrueFFS dramatically prolongs device lifetime by using advanced dynamic and static wear-leveling algorithms".
Meaning: even if the OS thinks its writing to the same sectors on disk, the flash system is smart enough to scatter repeated writes across the chip. So effectively you dont write to the same areas all the time. Check out their online disk life span calculator.
It doesn't matter if I move the queries to the database, that won't get rid of the initial delay in connecting to the database and executing the stored procedure or selecting the view. Getting a connection to oracle takes around.5 secs (on our server atleast). Besides, like I mentioned, there will be other hosts on the network which require the information from the database massaged into another format by my older functions. I will need a daemon to listen for those.
We have a host which contains configuration (in the database) and have clients that require this information. The problem with this is that the data models on each side are different, and the host side data has to be massaged to get it into the format acceptable to the client side. To make it more complicated; you can have _ALL_ the configuration sent to the clients at once, in which case the host side builds a virtual image of the client's contents in its memory and sends it all down to the client (takes around 100-150 queries to figure out what the client's image will look like). Or you can have a single update in which only 1 record on the host side gets updated; in which case doing 100-150 queries is inefficient. BUT If the server doesn't have a virtual map of the clients contents it doesn't know how to present the update to the client (one update can have an effect on multiple entries). I think I'm gonna put the logic on the client's end, the host will send the update and the client will process it and figure out what to do with it.
You are right, but was not the entire story. In my opinion, the daemon was required. The program opened a connection to the database and prepared 15-20 queries to execute. If it were not for the daemon, a new connection to the database + repeated preparation of the queries will need to be done _everytime_ the java calls were made. Also, this is a distributed system, clients from all over the network need access to this method.
Slashdot Mirror
5472 wasn't RC1 - it was a technical refresh of Beta 2.
RC1 is yet to come.
To clarify..
Kernel mode drivers need to be recompiled with each new release. Since the source of most drivers are already included in the kernel;s source tree this is not a problem.
Though, user-mode applications which everyone cares about have always been binary compatible with newer versions of the kernel. They only break if syscalls that the kernel exposes to user mode change, which they obviously have not. New syscalls are added with new releases of the kernel, but old ones are never removed or changed for compatibility reasons.
... then install windows.
2) start->control panel->network connections . right click properties on the adpater used for the internet connection.
3) Go to the Advanced tab and turn on the firewall.
4) Reconnect the machine to the network and start downloading the patches.
Screw the net ... get them to search my freakin hdd. Its only 80gigs.
It's MUCH harder to keep track of a changing hard disk with 50 gazillion file formats, compared to what google does with the (relatively) slow changing net and indexing pdf, ps and html files.
Lets get them Ph.Ds take a swag at that, eh?
Mod up parent!
How the fuck would you do that when your user base is stupid enough to double click an executable attached on an email from bob@growyourpenis.com?
Check out boa:
Boa currently seems to be the favorite web server in the embedded crowd, and embedded Linux, despite all the marketing hype, really is a big deal. Supposedly, an older version of Boa, v0.92q, runs in 32K address space on m68k, like used in uCLinux
The windows dragging feature is MUCH needed!
A Israel based research company is claiming to have created a computer a-i program that has passed the turning test for 15 month old baby: i.e when talking to it, you can not tell the difference between the computer and a 15 month old baby. By using a behaviorist approach (positive reinforemecent on desirable responses) they are planning to simulate and grow and adult mind in the next 10 years. The have video press releases here.
The slashdot heading is misleading; these guys are trying to make a system that will make it more expensive for mass emailers to send out spam. Not necessarily through money (although that is an option) but through exchange of resources, like cpu cycles. Very novel concept. I can imagine the same being done all over the internet. Instead of paying by cash you pay with something you have in abunadance (definately not cash) ... cpu cycles.
Suppose slashdot wants to make revenue: every user that hits the slashdot page agrees to donate some of their cpu resources to slashdot. slashdot inturn sells these resource for cash to GENOME xyz company which uses distributed computing to fold DNA sequences.
The patch for MSSQL vulnerability was out 6 months before the hammer virus hit. so that too was fixed "relatively" quickly.
as for half the internet not being slowed down by the vulnerability; maybe its because of this
Is this what you are taking about: here and here?
Both are under slashdot's recent apache news, btw.
you'd better be damn good to get an A, specially in Computer Science. The course averages for math and comp sci are nearly always in the C-,+ zones.
> no need to bother with sequence number guessing, just send your data packet right away, and pretend the connection is already open.
wrong. you still need to know the sequence number of the stream from the client to the server if you are going to send a packet. otherwise the server will drop the packet.
a friend of mine has black and white palm, he showed me a little program that turns the his palm into a mirror that he can use on the go.
>Too bad, though. Lack of competition will >decrease choice and progress. Intel can now slow
>down it's R&D and it may be longer before we see
>quantum chips and true AI. Worst of all, no
>realistic simulated customized porno. There goes
> my 3-breasted Klingon babe fantasy (sniff).
> I'll miss you AMD.
Now Intel can take all the R&D they had in making x86 compatiable chips and move it to something other, like quantum chips. They dont need to ensure their domination and have more felxiblity to explore other alternatives than tired x86 line.
I dont know how hardcore you have to be to dish out 45 bucks for an entire year of subscription to the Xbox live service. i'm wondering if Microsoft is losing money on this as well ...?
narrower flexible cables is all the SATA has to offer? its like switching from a two-headed screw driver standard to a philips one. the heads are different but the screw gose in the same way. give me a few features of SCSI in an affordable package and then your talking.
.. 2 hard disks, a dvd and cd writer later your computer would have enough ribbons to host a toddlers birthday party. and to get rid of the freaking master/slave shite ... ahhh. damn it I want SATA!
on the other hand: installing more ram/new-cpu wouldn't be such a pain
>Unfortunately, using Fat32 or NTFS(if you
>were "Gasp" running nt/2k), you would be
>repeatledy using the same flash sectors,
>quickly burning them out.
Thats why M-System's Disk on chip has it's TrueFFS file system, which "Prolongs Flash Life Span: TrueFFS dramatically prolongs device lifetime by using advanced dynamic and static wear-leveling algorithms".
Meaning: even if the OS thinks its writing to the same sectors on disk, the flash system is smart enough to scatter repeated writes across the chip. So effectively you dont write to the same areas all the time. Check out their online disk life span calculator.
Get the F out. Its WWE you nerds!
The Algorithm Design Manual by Steven S. Skiena.
It doesn't matter if I move the queries to the database, that won't get rid of the initial delay in connecting to the database and executing the stored procedure or selecting the view. Getting a connection to oracle takes around .5 secs (on our server atleast). Besides, like I mentioned, there will be other hosts on the network which require the information from the database massaged into another format by my older functions. I will need a daemon to listen for those.
We have a host which contains configuration (in the database) and have clients that require this information. The problem with this is that the data models on each side are different, and the host side data has to be massaged to get it into the format acceptable to the client side. To make it more complicated; you can have _ALL_ the configuration sent to the clients at once, in which case the host side builds a virtual image of the client's contents in its memory and sends it all down to the client (takes around 100-150 queries to figure out what the client's image will look like). Or you can have a single update in which only 1 record on the host side gets updated; in which case doing 100-150 queries is inefficient. BUT If the server doesn't have a virtual map of the clients contents it doesn't know how to present the update to the client (one update can have an effect on multiple entries). I think I'm gonna put the logic on the client's end, the host will send the update and the client will process it and figure out what to do with it.
You are right, but was not the entire story. In my opinion, the daemon was required. The program opened a connection to the database and prepared 15-20 queries to execute. If it were not for the daemon, a new connection to the database + repeated preparation of the queries will need to be done _everytime_ the java calls were made. Also, this is a distributed system, clients from all over the network need access to this method.
Intellectual laziness is never a virtue.
CORBA would have been overkill for the interface. It _is_ good, but designed for a vastly more distributed enviroment.
OpenCyc.org the open source cyc website and Cycorp the commercial website.