I know the feeling. I spend most of my time gathering requirements from clients (whether internal or external) and doing design before writing a single line of code. In some cases I might prototype something without a lot of design effort just to show that something is feasible.
After its all said and I have written the code it usually doesn't exceed an average of 10 lines per day for most short term projects I've worked on.
Measuring lines of code written per period of time has to be the worst metric for trying to evaluate the productivity of a developer. I would much rather have my worth determined by whether or not I meet my boss/self assigned objectives in a timely manner.
I wouldn't be surprised if the core of the NT 5.1 OS was the only thing in common between Windows XP Embedded and Windows XP. However, the fact that Microsoft does have an embedded version of XP that is not dependant on the GUI shell and IE to function proves that it is technically possible to not have IE present with the OS (which we always knew).
I don't believe for a moment Microsoft's claim that IE can't be removed from the current 2000/XP codebase. However, I'm going to give the company a break in this regard and say that I do believe it is probably technically infeasible (from a cost/manpower perspective) to go back and gut IE from the current 2000/XP codebases and deal with the resulting fallout of applications that depend on the core IE files breaking. I just have a problem in general with the idea that courts should be able to force any company (even Microsoft in this unfortunate instance) to write software in a particular manner.
I think some of the more effective remedies such as disclosure of all api's and prevention of exclusively licensing terms with OEM's (had they been levied by the courts) would have gone a long way towards restoring competiveness in the market and help prevent Microsoft from further abusing its monopoly power in the desktop OS market to conquer other markets.
Actually... there are boot time defraggers that will defrag the registry hive. Another simple thing to do is to install a minimalist parallel copy of NT/2000/XP (to recover from super nasty device driver failures) and defrag the drive from there (it will defrag the registry hive of the other install because it isn't in use).
Yes... you are telling me nothing that I didn't already know... I was just using an analogy between the P vs PPro and how it relates to the Itanium... I should have been more clear in my post...
So you are saying that AMD's 64-bit chip will be implemented as extensions to their current 32-bit architecture and not a complete redesign? Do you expect AMD's new processor to have issues with executing 32-bit code like Intel's PPro did with 16-bit code?
You must have me confused with someone else that doesn't look at price/performance/features/compatibility when making decisions about what hardware I purchase... as I currently write this message on a dual p3-1000 since AMD had no SMP solutions.
Isn't the Itanium (being a completely new core) really seeking to push code to 64-bit like the Pentium Pro did with 32-bit code on the server end? Sure the Pentium Pro had 16-bit backwards compatibility but it suffered horrible performance penalties when forced to run it. From what I understand the current benchmarks for 32-bit code executing on the Itanium aren't all that impressive. Has Intel had time to rationally design a decent set of extensions for the current P6 based processors that won't affect current 32-bit performance?
AMD's new chip OTOH seems to be designed with the goal of executing both 64-bit and 32-bit code with few performance penalties... seems like a no-brainer to me for who is going to capture the lowend and home user market at the very least.
The last time I read the AUP (a year ago) the wording was quite vague... vague enough that I figure if you became any sort of legal liability (ie: DMCA violations) or otherwise annoyed them (ie: taking up lots of the limited upstream bandwidth) they could cut you off claiming you violated the AUP by running a server...
In any case... I miss RR... *sniff* I'm on attbi and have a massive 128kbit/s upstream rate... not even good enough to host a decent AVP2 game...
Time Warner RR service in Austin specifically disallows running servers.
Hoewver, as I have to use my machine for development I had every conceivable type of server running at one point in time or another (IIS, Apache, etc).
The only network scans I saw from RR involved scanning for open smtp relay -- since mine wasn't open they never complained about it. During the CodeRed 1 & 2 debacle I was even kind enough to compile a list of IP addresses on their network that were scanning me from infected IIS boxes. They kindly reminded me that their AUP prohibited me from running servers but I told them that I have one machine for development and it is part of my job and there was nothing I could do about it. I also told them that there was no publicly accessible links/content being served and that I stayed on top of all patches/vulnerabilities. They seemed satisfied with that response.
For high school I would have to say that deploying DJGPP + RHIDE would be the least traumatic for your students.
I can understand that others might feel that "coddling" students with an IDE is a bad thing... but honestly -- if the "average" high school kids of today taking compu-sci related classes are anything like they were when I was in high school... they aren't going to get very far on a cygwin/vim/gcc/gdb type solution as the learning curve is far too steep for the limited time you have to actually teach in your typical high school course.
Your average student will end up spending more time asking questions along the lines of "How do I delete a block of text in VI?" as opposed to the more important questions centered around basic software development and debugging.
Although I really wish they wouldn't do away with the old sniper rifle.. it won't be nearly as bad as you think.
The sniper rifle from Tribes & Tribes 2 (large and open levels with smaller structures dispersed throughout) is laser based and reveals your location when you fire. Just because you see someone firing the sniper rifle doesn't mean that anyone from your team is in an immediate position to do anything about it.
From reading the preview I'm getting the idea that UT2 will be supporting similar style levels... and would therefore need a change to the sniper rifle to balance everything out.
I think he is probably referring to the GUI tools that come with the NT version of Oracle. Any DBA that is worth his/her salt is going to know how to use the sqlplus console.
Doesn't Oracle develop their new implementation of their database products on linux first now?
That is exactly the role I relegate any of my VB development to... GUI related and simple ADO stuff.
If I need custom controls I might be tempted to develop them in VB if they aren't going to be used extensively. Otherwise I tend to code any ActiveX/COM related stuff exclusively in C++ using MFC or ATL.
Several months ago Hotmail upgraded their spam filters. If you log into the hotmail website and crank the spam filter to its second highest filter level (the highest flags mail from anyone that isn't in your address book as spam) you will find that 90% or more of spam is correctly routed to your bulk mail folder. There is even an option to automatically delete messages that are routed to the bulk mail folder but I don't do that as some of the mailing/announcement lists I subscribe to will get flagged as spam and manually require me to add the email address they are sent from to the "not bulk mail" list.
The for-loop scoping bug has a command-line switch to toggle correct behavior. Unfortunately, with it on, large chunks of their own MFC code will no longer compile.
That is currently the case with VC++ 6 (and always will be) but I believe that issue has been resolved in VC++ 7.
Slashdot Mirror
I know the feeling. I spend most of my time gathering requirements from clients (whether internal or external) and doing design before writing a single line of code. In some cases I might prototype something without a lot of design effort just to show that something is feasible.
After its all said and I have written the code it usually doesn't exceed an average of 10 lines per day for most short term projects I've worked on.
Measuring lines of code written per period of time has to be the worst metric for trying to evaluate the productivity of a developer. I would much rather have my worth determined by whether or not I meet my boss/self assigned objectives in a timely manner.
Wouldn't this be possible if there was support for the UDF filesystem and packet writing? Is anybody work on it?
I wouldn't be surprised if the core of the NT 5.1 OS was the only thing in common between Windows XP Embedded and Windows XP. However, the fact that Microsoft does have an embedded version of XP that is not dependant on the GUI shell and IE to function proves that it is technically possible to not have IE present with the OS (which we always knew).
I don't believe for a moment Microsoft's claim that IE can't be removed from the current 2000/XP codebase. However, I'm going to give the company a break in this regard and say that I do believe it is probably technically infeasible (from a cost/manpower perspective) to go back and gut IE from the current 2000/XP codebases and deal with the resulting fallout of applications that depend on the core IE files breaking. I just have a problem in general with the idea that courts should be able to force any company (even Microsoft in this unfortunate instance) to write software in a particular manner.
I think some of the more effective remedies such as disclosure of all api's and prevention of exclusively licensing terms with OEM's (had they been levied by the courts) would have gone a long way towards restoring competiveness in the market and help prevent Microsoft from further abusing its monopoly power in the desktop OS market to conquer other markets.
Actually... there are boot time defraggers that will defrag the registry hive. Another simple thing to do is to install a minimalist parallel copy of NT/2000/XP (to recover from super nasty device driver failures) and defrag the drive from there (it will defrag the registry hive of the other install because it isn't in use).
Yes... you are telling me nothing that I didn't already know... I was just using an analogy between the P vs PPro and how it relates to the Itanium... I should have been more clear in my post...
So you are saying that AMD's 64-bit chip will be implemented as extensions to their current 32-bit architecture and not a complete redesign? Do you expect AMD's new processor to have issues with executing 32-bit code like Intel's PPro did with 16-bit code?
You must have me confused with someone else that doesn't look at price/performance/features/compatibility when making decisions about what hardware I purchase... as I currently write this message on a dual p3-1000 since AMD had no SMP solutions.
Isn't the Itanium (being a completely new core) really seeking to push code to 64-bit like the Pentium Pro did with 32-bit code on the server end? Sure the Pentium Pro had 16-bit backwards compatibility but it suffered horrible performance penalties when forced to run it. From what I understand the current benchmarks for 32-bit code executing on the Itanium aren't all that impressive. Has Intel had time to rationally design a decent set of extensions for the current P6 based processors that won't affect current 32-bit performance?
AMD's new chip OTOH seems to be designed with the goal of executing both 64-bit and 32-bit code with few performance penalties... seems like a no-brainer to me for who is going to capture the lowend and home user market at the very least.
The last time I read the AUP (a year ago) the wording was quite vague... vague enough that I figure if you became any sort of legal liability (ie: DMCA violations) or otherwise annoyed them (ie: taking up lots of the limited upstream bandwidth) they could cut you off claiming you violated the AUP by running a server...
In any case... I miss RR... *sniff* I'm on attbi and have a massive 128kbit/s upstream rate... not even good enough to host a decent AVP2 game...
Time Warner RR service in Austin specifically disallows running servers.
Hoewver, as I have to use my machine for development I had every conceivable type of server running at one point in time or another (IIS, Apache, etc).
The only network scans I saw from RR involved scanning for open smtp relay -- since mine wasn't open they never complained about it. During the CodeRed 1 & 2 debacle I was even kind enough to compile a list of IP addresses on their network that were scanning me from infected IIS boxes. They kindly reminded me that their AUP prohibited me from running servers but I told them that I have one machine for development and it is part of my job and there was nothing I could do about it. I also told them that there was no publicly accessible links/content being served and that I stayed on top of all patches/vulnerabilities. They seemed satisfied with that response.
For high school I would have to say that deploying DJGPP + RHIDE would be the least traumatic for your students.
I can understand that others might feel that "coddling" students with an IDE is a bad thing... but honestly -- if the "average" high school kids of today taking compu-sci related classes are anything like they were when I was in high school... they aren't going to get very far on a cygwin/vim/gcc/gdb type solution as the learning curve is far too steep for the limited time you have to actually teach in your typical high school course.
Your average student will end up spending more time asking questions along the lines of "How do I delete a block of text in VI?" as opposed to the more important questions centered around basic software development and debugging.
Just my $0.02...
Since lots of companies use Oracle Applications the answer is unfortunate yes....
Although I really wish they wouldn't do away with the old sniper rifle.. it won't be nearly as bad as you think.
The sniper rifle from Tribes & Tribes 2 (large and open levels with smaller structures dispersed throughout) is laser based and reveals your location when you fire. Just because you see someone firing the sniper rifle doesn't mean that anyone from your team is in an immediate position to do anything about it.
From reading the preview I'm getting the idea that UT2 will be supporting similar style levels... and would therefore need a change to the sniper rifle to balance everything out.
Perhaps its video card related (actually driver related)? Are you using the drivers that came on the XP cd?
The 64k size limit in notepad never applied to NT... only Windows 9x/ME...
I think he is probably referring to the GUI tools that come with the NT version of Oracle. Any DBA that is worth his/her salt is going to know how to use the sqlplus console.
Doesn't Oracle develop their new implementation of their database products on linux first now?
That is exactly the role I relegate any of my VB development to... GUI related and simple ADO stuff.
If I need custom controls I might be tempted to develop them in VB if they aren't going to be used extensively. Otherwise I tend to code any ActiveX/COM related stuff exclusively in C++ using MFC or ATL.
Several months ago Hotmail upgraded their spam filters. If you log into the hotmail website and crank the spam filter to its second highest filter level (the highest flags mail from anyone that isn't in your address book as spam) you will find that 90% or more of spam is correctly routed to your bulk mail folder. There is even an option to automatically delete messages that are routed to the bulk mail folder but I don't do that as some of the mailing/announcement lists I subscribe to will get flagged as spam and manually require me to add the email address they are sent from to the "not bulk mail" list.
If you are that desperate for a copy the standard version of MSVC goes for a whopping $99 USD.
I'm sure you can find copies even cheaper on ebay.
Doesn't Windows 2000 Terminal Server support both rdp4 and rdp5 clients?
Are you absolutely sure that worked for you? How does disabling a compiler warning affect the for loop scoping issue?
// conditional expression is constant
When I looked it up in MSDN I get the following:
#pragma warning(disable:4127)
The for-loop scoping bug has a command-line switch to toggle correct behavior. Unfortunately, with it on, large chunks of their own MFC code will no longer compile.
That is currently the case with VC++ 6 (and always will be) but I believe that issue has been resolved in VC++ 7.
One way to get around the various vendor differences in STL implementations is to use somethings like SGI's STL port on multiple platforms.
http://www.stlport.org
It isn't NT specific... I've never had problems (that weren't related to bad hardware) playing CD's under many different NT 4.0 systems.
If these are all the same type of system (ie: Dell, Compaq, etc) then I would see if the system manufacturer has any updates for the cdrom firmware.
The other route is to try and figure out who made the cdrom drive itself and see if they have any firmware updates for the drive.
ext3 works fine for me... so far... only 2 gigs of data though...
:)
XP is based on the NT (ie: NT 3.x, NT 4.0, 2000) codebase... Microsoft has finally thrown the 9x/ME codebase into the trash (so-to-speak).