As other posters have said, It doesn't matter much after you land your first job. Your transcript won't be looked at.
I do know that my employer has a very strong preference to those who have co-op experience.
After you land your job, it is what you accomplish that really matters - However, it may still make a difference - not that you will be choosen becuase of your school or your grades, but I think it may make a difference in what you produce and therefore your success. Many of our senior technology staff come from one of the top CS schools - I don't think that is a coincidence.
I suggest the value of search depends on what you use your computer for and maybe how good your memory is. To me it is tremendously important. Most of my job is spent managing information. Searching for things and adding things to my knowledge base. I love Google's Desktop search for this reason. I also use blinkx, but prefer the google UI implementation. Months or years after something was written, I need to go find it. I may not know if it was ppt, xls, doc, pdf, or xml. I may not have written. I may not have even read it, but I need to find it.
Increasingly I find that mult-media files matter. MP3 recordings of meetings, images of whiteboards, videos of presentatons are all fair game. (My hobbies include photography and genealogy. So findings pictures of people and places, correlated with GPS tracklogs is also of interest.)
I also find that, although I structure my file system according to something that makes sense initially, it won't be the way I want to search for it later. I usually file things accordning to client and project. But later I may need to 'find all system specifications where the DRP recovery time requirement was longer than 30 minutes'. This kind of search would require a lot of my time.
I need way better search, way better meta-data (which means system created because people don't), and more disk space;-).
When I first saw the spotlight demo, I was thrilled. A very small step to help me with my job.
Comparing the licensing and vendor support costs between MS and the new Sun offering, I roughly calculate Sun to be half the price of MS, assuming we upgrade MS products every four years.
The IT support cost estimates are very interesting.
This is based on rough estimates, I don't have the real numbers handy. We have about 20,000 seats.
Reviewers are reviewing books they actaully have bought and want to read. They have weeded out the high probability garbage, by the time they have made a purchase decision.
Things would be different if we saw reviews from/. readers of random samples of books. But that would likely be less useful.
This is good stuff. Although the experiment used physical access to stress the memory, the theory could be used as an exploit in real situations in ways that the narrow of mind (like me) cannot conceive.
Perhaps this is not a method of practical attack on a machine. But it may be just a matter of creative thinking.
The key take away is to not disallow the possiblity.
Threats you discard as harmless is a logical place for an attacker to begin. Remeber the
Maginot line.
We have a relatively large development shop. About 600 developers. We spend tons of money on commercial software.
The quality of support we get varies greatly from vendor to vendor.
Almost always you have to recreate the problem in order to gather the needed diagnostics or to be able to test the fix. This takes time. It may be the same for OSS and COTS.
However, more often tha I can count, we have debugged the problem in the vendors code and sent them the fix! Why? We need the fix in a hurray, so we zap our install. We sent them the fix to get it as part of the 'official' code base. Combine that with the previous point and OSS may win.
We had another case where the COTS code was difficult to deal with, sucked CPU, and leaked memory like a sieve. We dumped it and replaced it with an OSS equivalent. It worked like a charm - fast and tight. It has cost us almost nothing to use. If it breaks and can't fix it, we can always revert to the COTS. And we'll be ahead.
Our policy is that we can use OSS where we feel appropriate, but we must obtain an internal owner for ongoing support. We have done well with it so far and will use more.
One thing to watch out for is the licensing obligations with some of the OSS agreements.
Slashdot Mirror
Shows you the streets.
For S390 based host systems (IMS, CICS) there are J2C adapters. IBM and 3rd party.
Of course we have no idea if you need such a thing for your product.
As other posters have said, It doesn't matter much after you land your first job. Your transcript won't be looked at.
I do know that my employer has a very strong preference to those who have co-op experience.
After you land your job, it is what you accomplish that really matters - However, it may still make a difference - not that you will be choosen becuase of your school or your grades, but I think it may make a difference in what you produce and therefore your success. Many of our senior technology staff come from one of the top CS schools - I don't think that is a coincidence.
I suggest the value of search depends on what you use your computer for and maybe how good your memory is. To me it is tremendously important. Most of my job is spent managing information. Searching for things and adding things to my knowledge base. I love Google's Desktop search for this reason. I also use blinkx, but prefer the google UI implementation. Months or years after something was written, I need to go find it. I may not know if it was ppt, xls, doc, pdf, or xml. I may not have written. I may not have even read it, but I need to find it.
;-).
Increasingly I find that mult-media files matter. MP3 recordings of meetings, images of whiteboards, videos of presentatons are all fair game. (My hobbies include photography and genealogy. So findings pictures of people and places, correlated with GPS tracklogs is also of interest.)
I also find that, although I structure my file system according to something that makes sense initially, it won't be the way I want to search for it later. I usually file things accordning to client and project. But later I may need to 'find all system specifications where the DRP recovery time requirement was longer than 30 minutes'. This kind of search would require a lot of my time.
I need way better search, way better meta-data (which means system created because people don't), and more disk space
When I first saw the spotlight demo, I was thrilled. A very small step to help me with my job.
Geotag my images from my GPS data, publish via picasa - sweet.
Or, point to place on map with keyhole, query local images using google desktop to find those close by?
A service not software, but does the job.
Comparing the licensing and vendor support costs between MS and the new Sun offering, I roughly calculate Sun to be half the price of MS, assuming we upgrade MS products every four years.
The IT support cost estimates are very interesting.
This is based on rough estimates, I don't have the real numbers handy. We have about 20,000 seats.
Reviewers are reviewing books they actaully have bought and want to read. They have weeded out the high probability garbage, by the time they have made a purchase decision.
/. readers of random samples of books. But that would likely be less useful.
Things would be different if we saw reviews from
From an english langauge pov:
http://dictionary.reference.com/search?q=fibre
From a google pov:
"fibre channel array" hit count = 2600
"fiber channel array" hit count = 213
This is good stuff. Although the experiment used physical access to stress the memory, the theory could be used as an exploit in real situations in ways that the narrow of mind (like me) cannot conceive.
Perhaps this is not a method of practical attack on a machine. But it may be just a matter of creative thinking.
The key take away is to not disallow the possiblity.
Threats you discard as harmless is a logical place for an attacker to begin. Remeber the Maginot line.
We have a relatively large development shop. About 600 developers. We spend tons of money on commercial software.
The quality of support we get varies greatly from vendor to vendor.
Almost always you have to recreate the problem in order to gather the needed diagnostics or to be able to test the fix. This takes time. It may be the same for OSS and COTS.
However, more often tha I can count, we have debugged the problem in the vendors code and sent them the fix! Why? We need the fix in a hurray, so we zap our install. We sent them the fix to get it as part of the 'official' code base. Combine that with the previous point and OSS may win.
We had another case where the COTS code was difficult to deal with, sucked CPU, and leaked memory like a sieve. We dumped it and replaced it with an OSS equivalent. It worked like a charm - fast and tight. It has cost us almost nothing to use. If it breaks and can't fix it, we can always revert to the COTS. And we'll be ahead.
Our policy is that we can use OSS where we feel appropriate, but we must obtain an internal owner for ongoing support. We have done well with it so far and will use more.
One thing to watch out for is the licensing obligations with some of the OSS agreements.