Yup, I currently develop software in that scale... I am doing "volume testing" right now so I have two "sandboxes" to work in. 1 16xDual-core Solaris machine with Oracle database shared on same hardware, and 1 48core IBM (SMT core - like Pentium HT - looks like 96 CPU) p595 which is partitioned.... I have 8 cores for "my" DB2, 16 cores for "me", and someone else plays with the rest....
For our application these machines are over-spec'd. While our app has many components in many languages, (COBOL, C, Java, Perl), I am responsible for some of the Java parts.... and... well, scalability in the primary Java component is linear to a point, but then there is some other bottleneck....... in our case, the database (or, more accurately, the disks supplying data to the database... actually, it really is the Fiber-channel SAN infrastructure limiting me to two 2Gigabit connections to the disks).
So, we are linnear in both enviroments AIX+db2 and Solaris+Oracle to about 10 to 12 CPU's, but then start hitting data starvation at the database end.
I imagine that every "enterprise" encounters a limit somewhere...
gus.
Now, if only I could get another 2 HBAs and one more DS4800's, then I could probably scale through to about 20 CPU's but that would cost 200K.
What I have: "breadboard" for playing with circuits. I found DIY Kit 128 at a local electronics store. The website is at http://www.kitsrus.com/upuc.html The Kit128 is undergoing some support problems right now in that the primary software developer died, and there is a new crowd taking over. It still works really well, but the newer chips will be a few weeks away from being supported. See discussions at: http://www.websitetoolbox.com/tool/post/diykit/vpo st?id=760548&trail=30
If I were starting over I would consider one of the EasyPROG, ICD2, Wisp628, or the KIT128/149/150/182 and I believe the 185 (in no particular order). It looks like the Kit128 and it's siblings still have a lot of life in them yet. There are other programmers, and everyone has a different opinion as to whether you should build your first programmer, or buy it. I got the 128 as a kit, and was happy.
After that, everything you need is available for free from Microchip itself. The IDE is good, ASM is fun, there is even a free "student edition" C Compiler, and you can get free samples of the actual chips themselves.
In something of the same spot myself. I have found great satisfaction programming PIC Microcontrollers. Recently I decided to move from assembler to C, and that is quite fun too. It's amazing what you can put together when you try.
The answer to archiving the required volumes is producing less volumes. Case in point... we recently spent a week or so at work optimising a process that was I/O bound. The bugger took 10 hours to run. Although purchasing faster disks, converting to RAID0, and other techniques did whittle down the execution time to about 5 hours, the final solution was to redefine the process to reduce the actual IO (removed a COBOL sorting stage in the process), and the process is now 2 hours.
Bottom line: with the 100 + 38 million dollars (FTFA) assigned to the project I am sure I could eliminate a number of redundant positions, optimise some communication channels, retire voluminous individuals, replace inefficient protocols/people, and basically reduce the sources of data. Hell, if the US were to actually have peace instead of demand it, there would be a much reduced need for military inteligence, political rhetoric, and other civil responsibilities. The military could be half the size, and what do you know, we could not only reduce the requirement for archiving, but could actually save money in the process.
put it on a geared mechanism, and then hang it high in your container. get a bio-degradable material, or some other perishable material that takes about 1000 years to degrade.
Use the degrade material to suspend the weight.
After 1000 years the material fails, releasing the weight. The gearing mechanism is used, like the mechanism on a grandfather clock that powers the pendulum, to "tick" every 10 seconds. At each 10 second interval a large ball-bearing will be dropped on to the now-defunct cymbal from your 80's drum set.
This audio signal will be far more likely discovered than a RF type signal.
Your only limits are how heavy the weight can be, how many ball-bearings you can trap at the top of your container, and how deep you bury it.
I have found that some formats are far better at some data types than others.
e.g. for: text,ascii,documents: use any of bzip2, gzip, zip. audio: use nothing if MP3/AAC/etc. flac for other "raw" formats video: use the most appropriate encoding (mpeg4/divx,etc) and then don't try to compress.
bzip encodes/decodes slower, but has typically better compression ratios.
So, use whatever people commonly use for the data type you are compressing.
My initial reaction to the quote (remember, I am a Linux bigot), is that Microsoft needs thousands more active developers just to maintain the convoluted code, and to fix bugs. The Linux code is that much cleaner, and has fewer requirements for keeping legacy backward-compatible code.
i.e. Microsoft is burdened by compatibility and legacy issues. Linux shrugs off the bad habits, and moves on.
The result is that Linux has comparable "active" functionality, but a miniscule amount of "legacy" functionality, and thus there is simply less work to do for the developers. Also, the code is simpler, and, frankly, the intellectual demands on a Linux developer are less.... so, an intelligent linux developer can spend more of their brain cycles conceptualising improvements instead of finding work-arounds.
I have lots of work experience in environments where logacy code abounds, and I have spent lots of time working on new and emerging systems as well. My experience indicates that most of the issues in legacy systems involves regression, but the baggage-free new tools are unencumbered, and thus have more scope for "fun" enhancements.
If Linux (and open source in general) were to make commitments to backward compatibility I am sure that the developers would quickly become entangled in maintenance, rather than development.
Bottom line is that the OP Quote is accurate in the sense that Microsoft has an encumbered product, Linux is free of those restrictions, and is thus leaner (code wie, and developer wise).
Linus himself is recently quoted as saying that the major push in linux is no longer in the kernel, it is user-space... implying that there is no more real "sexy" stuff to do there.
Hmmmm... good thing you qualified your post with IANAL....
Actually, to answer your questions. Yes, you will have to pay them if they (fairly) patented the algorithm and have demanded payment (as has always been the case).
Also, Open source has always been vulnerable to the "Hey the open source.... infringes on my patent...." scenario just like all other software sources other than the open one.
As for comming up with an alternate algorithm to solve a problem for which there already is a patented algorithm, well, this happens all the time, and is not worthy of discussion (and no, the "big company" could not "take you out", at least not in theory.... not for using an algorithm they did not patent)
About the only thing interesting about your post is your Sig, and even that is immature.
R/C has licensed a number of the 72MHz bands (for planes) for years (decades, actually). There is no leakage from R/C TX's into other bands, and BPL is as much a threat to R/C as other uses. BPL has a broad spectrum emmission, RC has a valid, licensed, and exclusive right (with the FCC) to the frequencies they use. i.e. The ONLY people generating RF noise (for non-RC purposes) in RC frequencies are doing so illegally.
Thus, instead of trying to sound like you know what you are saying, get your facts straight. Since you know what QRM is, you should also know that R/C is fairly allocated it's frequencies.
Hmmmm.. I said it in jest, but there is a sentiment like it in EU. I am a South African lived in UK for 3 years, and now live in Canada.
On the whole, Americans have a negative reputation. Canadians are mostly welcomed. A Canadians biggest fear is that people will assume they are americans, and thus they prominantly display a Canadian flag somewhere.... especially back-packers.
I doubt an American will get a job in france if there is a canadian applying as well, with similar skills. My perception is that is common accross the EU.
So, I am not totally BS'ing, but there is jest in there. As I say though, you may bet your foot in the door by lying, but you will get booted out promptly when they discover the lie.
Slashdot Mirror
I know, it is jut a typo...
(from page: http://www-304.ibm.com/jct09002c/university/stude
gus
Well, I am somewhat surprised. I use Firefox 1.5.0.3, and it does not show the "Slashdot" "News for nerds" in the linked "preview".
/. is going under cover. I get a link if I over over where the /. title should be.
I guess
Go figure...
gus
Yup, I currently develop software in that scale... I am doing "volume testing" right now so I have two "sandboxes" to work in. 1 16xDual-core Solaris machine with Oracle database shared on same hardware, and 1 48core IBM (SMT core - like Pentium HT - looks like 96 CPU) p595 which is partitioned .... I have 8 cores for "my" DB2, 16 cores for "me", and someone else plays with the rest....
... in our case, the database (or, more accurately, the disks supplying data to the database ... actually, it really is the Fiber-channel SAN infrastructure limiting me to two 2Gigabit connections to the disks).
For our application these machines are over-spec'd. While our app has many components in many languages, (COBOL, C, Java, Perl), I am responsible for some of the Java parts.... and... well, scalability in the primary Java component is linear to a point, but then there is some other bottleneck....
So, we are linnear in both enviroments AIX+db2 and Solaris+Oracle to about 10 to 12 CPU's, but then start hitting data starvation at the database end.
I imagine that every "enterprise" encounters a limit somewhere...
gus.
Now, if only I could get another 2 HBAs and one more DS4800's, then I could probably scale through to about 20 CPU's but that would cost 200K.
Well, I am still mostly a beginner myself.
o st?id=760548&trail=30
What I have:
"breadboard" for playing with circuits.
I found DIY Kit 128 at a local electronics store. The website is at http://www.kitsrus.com/upuc.html
The Kit128 is undergoing some support problems right now in that the primary software developer died, and there is a new crowd taking over. It still works really well, but the newer chips will be a few weeks away from being supported. See discussions at: http://www.websitetoolbox.com/tool/post/diykit/vp
If I were starting over I would consider one of the EasyPROG, ICD2, Wisp628, or the KIT128/149/150/182 and I believe the 185 (in no particular order). It looks like the Kit128 and it's siblings still have a lot of life in them yet. There are other programmers, and everyone has a different opinion as to whether you should build your first programmer, or buy it. I got the 128 as a kit, and was happy.
After that, everything you need is available for free from Microchip itself. The IDE is good, ASM is fun, there is even a free "student edition" C Compiler, and you can get free samples of the actual chips themselves.
Be sure to visit the microchip forums, piclist.com, and http://www.voti.nl/pic/index.html
enjoy. I do.
gus
In something of the same spot myself. I have found great satisfaction programming PIC Microcontrollers. Recently I decided to move from assembler to C, and that is quite fun too. It's amazing what you can put together when you try.
www.microchip.com
www.piclist.com
gus
Uhmmm.. It did not pass.
gus
... very easy.
I'm now on 1.5RC1
Thanks
gus
You get additional karma if you get her to read the comment you just posted ....
For $5 in my paypal I'll drop her an anonymous e-mail to point her in the right direction.
gus
Assuming parent is a bona-fide posting ... it scares me that the developers of airplane navigation systems are content with "quite stable".
gus
I believe the regulatory body that would be concerned about this is the CRTC (Canada Radio-television and Telecommunications Commission).
I have lodged a complaint with them at:
http://www.crtc.gc.ca/
Feel free to do the same.
gus
Don't hold your breath....
gus
Of course, you could use eclipse .... see www.eclipse.org/ve/
gus
People should think outside the box.
The answer to archiving the required volumes is producing less volumes. Case in point... we recently spent a week or so at work optimising a process that was I/O bound. The bugger took 10 hours to run. Although purchasing faster disks, converting to RAID0, and other techniques did whittle down the execution time to about 5 hours, the final solution was to redefine the process to reduce the actual IO (removed a COBOL sorting stage in the process), and the process is now 2 hours.
Bottom line: with the 100 + 38 million dollars (FTFA) assigned to the project I am sure I could eliminate a number of redundant positions, optimise some communication channels, retire voluminous individuals, replace inefficient protocols/people, and basically reduce the sources of data. Hell, if the US were to actually have peace instead of demand it, there would be a much reduced need for military inteligence, political rhetoric, and other civil responsibilities. The military could be half the size, and what do you know, we could not only reduce the requirement for archiving, but could actually save money in the process.
Remeber, govenment is a self-supporting process.
Go ahead, mark me a troll.
gus
Man you have some steep requirements.... eh!
The market here is none-too-rosy, but for your peimary needs look first in the Subject areas.
gus
Actually, it was a brain fart .... oops. I did mean Avro.
Although Avril is not too shabby either.
gus
CBC Archives - see "did you know"
gus
Gutenburg
http://www.gutenberg.org/
gus
....
;-)
I did something like this once....
put it on a geared mechanism, and then hang it high in your container. get a bio-degradable material, or some other perishable material that takes about 1000 years to degrade.
Use the degrade material to suspend the weight.
After 1000 years the material fails, releasing the weight. The gearing mechanism is used, like the mechanism on a grandfather clock that powers the pendulum, to "tick" every 10 seconds. At each 10 second interval a large ball-bearing will be dropped on to the now-defunct cymbal from your 80's drum set.
This audio signal will be far more likely discovered than a RF type signal.
Your only limits are how heavy the weight can be, how many ball-bearings you can trap at the top of your container, and how deep you bury it.
YMMV
gus
I refuse to comment on posts that do not include a link that can be slashdotted!
Hang on, that makes this an oxy-moron....
gus
I have found that some formats are far better at some data types than others.
e.g. for:
text,ascii,documents: use any of bzip2, gzip, zip.
audio: use nothing if MP3/AAC/etc. flac for other "raw" formats
video: use the most appropriate encoding (mpeg4/divx,etc) and then don't try to compress.
bzip encodes/decodes slower, but has typically better compression ratios.
So, use whatever people commonly use for the data type you are compressing.
gus
My initial reaction to the quote (remember, I am a Linux bigot), is that Microsoft needs thousands more active developers just to maintain the convoluted code, and to fix bugs. The Linux code is that much cleaner, and has fewer requirements for keeping legacy backward-compatible code.
i.e. Microsoft is burdened by compatibility and legacy issues. Linux shrugs off the bad habits, and moves on.
The result is that Linux has comparable "active" functionality, but a miniscule amount of "legacy" functionality, and thus there is simply less work to do for the developers. Also, the code is simpler, and, frankly, the intellectual demands on a Linux developer are less.... so, an intelligent linux developer can spend more of their brain cycles conceptualising improvements instead of finding work-arounds.
I have lots of work experience in environments where logacy code abounds, and I have spent lots of time working on new and emerging systems as well. My experience indicates that most of the issues in legacy systems involves regression, but the baggage-free new tools are unencumbered, and thus have more scope for "fun" enhancements.
If Linux (and open source in general) were to make commitments to backward compatibility I am sure that the developers would quickly become entangled in maintenance, rather than development.
Bottom line is that the OP Quote is accurate in the sense that Microsoft has an encumbered product, Linux is free of those restrictions, and is thus leaner (code wie, and developer wise).
Linus himself is recently quoted as saying that the major push in linux is no longer in the kernel, it is user-space... implying that there is no more real "sexy" stuff to do there.
Food for thought
gus
Hmmmm... good thing you qualified your post with IANAL....
.... infringes on my patent...." scenario just like all other software sources other than the open one.
Actually, to answer your questions. Yes, you will have to pay them if they (fairly) patented the algorithm and have demanded payment (as has always been the case).
Also, Open source has always been vulnerable to the "Hey the open source
As for comming up with an alternate algorithm to solve a problem for which there already is a patented algorithm, well, this happens all the time, and is not worthy of discussion (and no, the "big company" could not "take you out", at least not in theory.... not for using an algorithm they did not patent)
About the only thing interesting about your post is your Sig, and even that is immature.
Hmmm....
gus
This is shuch BS.
R/C has licensed a number of the 72MHz bands (for planes) for years (decades, actually). There is no leakage from R/C TX's into other bands, and BPL is as much a threat to R/C as other uses. BPL has a broad spectrum emmission, RC has a valid, licensed, and exclusive right (with the FCC) to the frequencies they use. i.e. The ONLY people generating RF noise (for non-RC purposes) in RC frequencies are doing so illegally.
Thus, instead of trying to sound like you know what you are saying, get your facts straight. Since you know what QRM is, you should also know that R/C is fairly allocated it's frequencies.
gus
Hmm ... I fly R/C regularly. The submitter doesn;t actually say much.... but his intended use for this is simply wrong.
This is a Transmitter, a TX. Not an RX (Receiver). There will be no telemetry to the screen from the plane.
gus
Hmmmm.. I said it in jest, but there is a sentiment like it in EU. I am a South African lived in UK for 3 years, and now live in Canada.
On the whole, Americans have a negative reputation. Canadians are mostly welcomed. A Canadians biggest fear is that people will assume they are americans, and thus they prominantly display a Canadian flag somewhere.... especially back-packers.
I doubt an American will get a job in france if there is a canadian applying as well, with similar skills. My perception is that is common accross the EU.
So, I am not totally BS'ing, but there is jest in there. As I say though, you may bet your foot in the door by lying, but you will get booted out promptly when they discover the lie.
gus