End users should never, EVER have to recompile a kernel. Linux distros ship highly modular kernels that work with pretty much anything. Although slight performance or efficiency improvements can be had by compiling specifically for your hardware, for the most part it just isn't something that will be worth doing for the kind of user you describe.
Seriously, why SHOULDN'T you do this? The only thing Verisign does is take exorbitant amounts of money to "prove" you are who you say you are. But if you don't trust someone at their word, you probably don't want to do business with them in the first place!
I'd suggest that doing this even for sites used by the general public is OK. Just put a quick explanation on the site. The exception might be if you're running a large operation collecting credit card numbers, in which case you can afford Veri$ign's price and don't want to lose a bit of business.
For what it's worth, I just got my old PHP scripts running under Psyche's Apache+PHP. I just set the PHP options register_globals and short_open_tag to ON. Haven't done extensive testing yet, but anecdotally, the script runs, and even connects to my Postgres database.
If the language was a bit cleaner, I'd recommend to mod parent up. Installing Apache 1.3 is no big deal. You can probably even easily rebuild the 1.3 RPMs if you prefer to keep everything in that format.
Red Hat may have jumped the gun by including Apache 2 before PHP and mod_perl are proven on it, but in the lang run it should help to iron out the bugs. Just like what happened when they were the first to include glibc 2.0 in Red Hat 5.0.
In the meantime, if 2.0 doesn't work for you, just use 1.3! Any Linux admin worth $5/hour will be able to set that up.
I've been a KDE user for years. Logged into Gnome 2.0, figured out the keybindings that were different from KDE, and I *still* haven't even logged into KDE since installing Psyche! I think I can live with this. I'll try Psyche's KDE, so whether I stick with Gnome remains to be seen.
Attempted to switch to Evolution. The SECOND freeking e-mail I received crashed the mail component of Evolution every time I clicked on it! I was like "uhh, no." Back to Kmail.
Started KDevelop. Told it to index the KDE/Qt documentation on setup. The ht:/dig process just kept going and going and going, spidering over EVERYTHING on my freeking filesystem! I eventually had to kill it.
Maybe I should have stuck with 1024X768 resolution instead of upping it to 1280X1024. It's a tad flickery, but not bad. I have a G400 and a 19" monitor.
Other than that things are working fine.
Re:Why Red Hat won't beat windows on the desktop
on
Red Hat 8.0 Reviewed
·
· Score: 2
Have you seen GNU Enterprise? It is aiming to fill that market segment. Granted, it's probably a couple years away from being a serious competitor, but it IS being used in some businesses TODAY.
I also believe the big ERP/CRM people will put out Linux clients at some point. A lot of businesses are dying to dump Microsoft. Oracle would certainly like to help them do that. SAP also has some Linux support on the server, so they're probably open to the idea. We'll just have to wait and see.
And it's a great book! I'm spending this week inhaling it and practicing the stuff.
I submitted it as a story to Slashdot and it was rejected.:/ This book is a lot more important than most of the other recent stories on here. Mozilla has WAY more power than you ever dreamed, and this book tells you how to use it. Highly recommended.
Agree wholeheartedly with the first reply to parent.
XUL makes it possible to quickly develop cross-platform applications that load like web pages but look like "regular" desktop apps.
That's great news. I for one am tired of using applications that are done with just HTML. It's not what HTML is designed for, and we need something better. XUL provides that.
It's also a potent weapon we can use against IE. I'm convinced that we're in a very dangerous situation right now. If Microsoft can get some of the bigger sites to only work with IE, you can kiss goodbye all hopes for competition in the web browser and operating system market. With its current market share, we're dangerously close to that level. The solution, of course, is to get people to use Mozilla!
And why would end users care about switching to Mozilla? APPLICATIONS!
For this reason, I advocate doing new Web development work in XUL instead of HTML. Not only does it look MUCH nicer than traditional web apps, but it will give people a reason to switch to Mozilla.
I'm currently inhaling O'Reilly's new Mozilla application book. It's available under an Open Content license. (I submitted this as a story to Slashdot but they rejected it!!! Why??? This is HUGE!) The book is a good one and it can really show you what Mozilla is capable of. It is a very slick environment. Please check it out!
Hmm I was in Ecuador for a month earlier this year (mostly working at a school in Quito but also traveling around in the Andes region to Riobamba) and I don't think I ever saw a power outage! I don't think most of the PCs at the school had UPSes and we never had trouble.
I *have* seen that in other countries -- Honduras has HORRIBLE power, even the big cities. But I haven't been there since 1994, so maybe things have improved.
Granted that this is just gut instinct, but I really don't think that would fly. By any reasonable definition, a beta tester is NOT part of the company. In fact, going back to the original definitions, alpha testing is done inside the company and beta testing is done outside the company, once the product is deemed to be in somewhat stable shape internally.
The real question is whether or not a closed beta does indeed qualify the participant as an agent of the company.
I would guess that it does not. A closed beta recipient is (generally) not being paid by the company, has no permanent relationship, is installing the software himself, and is not using the company's hardware. IANAL.
However, UL might just be intending to apply the NDA to the non-GPL parts of the distro. Why would they care if someone took the gcc from their closed beta and redistributed it? The installer and config utilities are what they're concerned about (I think). Of course, if they have a heavily modified KDE, then they would NOT be able to keep their beta testers from redistributing that, including their modifications.
interesting. I was thinking of downloading knoppix to burn and give out to people. I have an HP internal 8x IDE burner (8100i or something like that). Got it in summer 2000. Would that be too old and unable to burn it?
(Lest someone just say "download it and try it", it would be nice to save the bandwidth if it won't work!)
Simply require the user to start browsing at a web page, which of course lists "special deals" at local businesses. Make them click on at least one for more info. Then just let them surf for a while with no more interruptions.
Every time a new DHCP lease is assigned, make them go to this page and click an ad. Make the leases last maybe 2-4 hours.
I don't think that would annoy anyone too badly. The question is could you get enough interest from local businesses to fund the thing and feed yourself. Heck, if that would be profitable I might try to set something like that up here in Salem!
Now throw in plugins, so for instance the FS layer interprets JPEGs and adds extra attributes. Now you can read the colour depth of an image by doing "cat photo.jpg/colour_depth" or whatever. You can get the raw, uncompressed version of the file by doing "cp photo.jpg/raw > photo.raw".
Yikes! And this belongs in the kernel???
And even if it does belong there, wouldn't it be best in the general FS layer than in a specific FS? It could work across ALL filesystems...
Actually, even though the Linux market is much smaller, there's less competition with other vendors there, so if you produce a quality program, you're pretty much guaranteed to get a good number of sales.
I can think of several applications that would probably sell well on Linux... lots of edutainment type programs (Oregon Trail, etc), a Print-Shop type program that would easily create greeting cards and banners and posters, a good personal finance program (which TheKompany apparently has in Kapital), a QuickBooks workalike that really is as easy as QB...
If ALL that software was available and hardware companies offered it as a preload option, seriously, why would more people not consider Linux?
(1) Some unexpected people are going to _need_ MS Office, point-blank. The popular financial package we use only 'exports' data to Excel. Not excel file formats; just Excel (via OLE or COM or whatever they're calling it this week.) Although it has a 'print to CSV' feature (don't ask), it saves the file with some silly Lotus-specific extension that OpenOffice doesn't believe is actually a CSV file. Although renaming should in principle be easy, the people who need to use this data are simply not up to the task of understanding (a) why they need to set their folder options to show all those funny little three-letter thingies at the end of the filename, and (b) why they can't just click the 'Excel' button like they used to. So the people that need to regularly manipulate the data in the financial database at a relatively low level need MSO.
Crimony, the solution to this is so easy it isn't funny. Write an OpenOffice Basic macro to look for the CSV file, rename it if necessary, get OOo to load it, maybe even automatically apply some custom formatting. You could even have it load all the CSVs at once if there are more than one.
Want to hire me to implement that?:D I'm learning OOo macro programming and need some paying work....
hmm, granted I'm no power user at this point, but I'm pretty sure all that is possible. You might want to ask on the OpenOffice users list... seriously...
I agree that the documentation leaves a bit to be desired, but they ARE working on it.
For one thing, there's a new Software Development Kit with info on how to write plugins.
Here's a link to thew new developer documentation draft.
OOo lets you put a Basic macro in a document and have it run when the doc is openned. Couldn't it, say, look around in your home directory for text files or OOo documents, load said documents, and "hide" interesting info from those documents in the current document?
End users should never, EVER have to recompile a kernel. Linux distros ship highly modular kernels that work with pretty much anything. Although slight performance or efficiency improvements can be had by compiling specifically for your hardware, for the most part it just isn't something that will be worth doing for the kind of user you describe.
> Plus I wouldn't be surprised to see some code like if(isMicrosoftSoftware) dontSwapOut(); down in the bowels of Windows somewhere.
Whoops. I think you mean:
IF isMicrosoftSoftware THEN
dontSwapOut
END IF
Seriously, why SHOULDN'T you do this? The only thing Verisign does is take exorbitant amounts of money to "prove" you are who you say you are. But if you don't trust someone at their word, you probably don't want to do business with them in the first place!
I'd suggest that doing this even for sites used by the general public is OK. Just put a quick explanation on the site. The exception might be if you're running a large operation collecting credit card numbers, in which case you can afford Veri$ign's price and don't want to lose a bit of business.
For what it's worth, I just got my old PHP scripts running under Psyche's Apache+PHP. I just set the PHP options register_globals and short_open_tag to ON. Haven't done extensive testing yet, but anecdotally, the script runs, and even connects to my Postgres database.
If the language was a bit cleaner, I'd recommend to mod parent up. Installing Apache 1.3 is no big deal. You can probably even easily rebuild the 1.3 RPMs if you prefer to keep everything in that format.
Red Hat may have jumped the gun by including Apache 2 before PHP and mod_perl are proven on it, but in the lang run it should help to iron out the bugs. Just like what happened when they were the first to include glibc 2.0 in Red Hat 5.0.
In the meantime, if 2.0 doesn't work for you, just use 1.3! Any Linux admin worth $5/hour will be able to set that up.
Personally I *like* the way PHP registers global variables from POSt and GET queries. Makes it really easy to code in.
Just make sure you initialize all your variables and you don't have a problem! I'll probably set that option to ON.
Installed it yesterday.
I've been a KDE user for years. Logged into Gnome 2.0, figured out the keybindings that were different from KDE, and I *still* haven't even logged into KDE since installing Psyche! I think I can live with this. I'll try Psyche's KDE, so whether I stick with Gnome remains to be seen.
Attempted to switch to Evolution. The SECOND freeking e-mail I received crashed the mail component of Evolution every time I clicked on it! I was like "uhh, no." Back to Kmail.
Started KDevelop. Told it to index the KDE/Qt documentation on setup. The ht:/dig process just kept going and going and going, spidering over EVERYTHING on my freeking filesystem! I eventually had to kill it.
Maybe I should have stuck with 1024X768 resolution instead of upping it to 1280X1024. It's a tad flickery, but not bad. I have a G400 and a 19" monitor.
Other than that things are working fine.
Have you seen GNU Enterprise? It is aiming to fill that market segment. Granted, it's probably a couple years away from being a serious competitor, but it IS being used in some businesses TODAY.
I also believe the big ERP/CRM people will put out Linux clients at some point. A lot of businesses are dying to dump Microsoft. Oracle would certainly like to help them do that. SAP also has some Linux support on the server, so they're probably open to the idea. We'll just have to wait and see.
Anyone know if that could work with Kylix on Linux?
And it's a great book! I'm spending this week inhaling it and practicing the stuff.
:/ This book is a lot more important than most of the other recent stories on here. Mozilla has WAY more power than you ever dreamed, and this book tells you how to use it. Highly recommended.
I submitted it as a story to Slashdot and it was rejected.
Agree wholeheartedly with the first reply to parent.
XUL makes it possible to quickly develop cross-platform applications that load like web pages but look like "regular" desktop apps.
That's great news. I for one am tired of using applications that are done with just HTML. It's not what HTML is designed for, and we need something better. XUL provides that.
It's also a potent weapon we can use against IE. I'm convinced that we're in a very dangerous situation right now. If Microsoft can get some of the bigger sites to only work with IE, you can kiss goodbye all hopes for competition in the web browser and operating system market. With its current market share, we're dangerously close to that level. The solution, of course, is to get people to use Mozilla!
And why would end users care about switching to Mozilla? APPLICATIONS!
For this reason, I advocate doing new Web development work in XUL instead of HTML. Not only does it look MUCH nicer than traditional web apps, but it will give people a reason to switch to Mozilla.
I'm currently inhaling O'Reilly's new Mozilla application book. It's available under an Open Content license. (I submitted this as a story to Slashdot but they rejected it!!! Why??? This is HUGE!) The book is a good one and it can really show you what Mozilla is capable of. It is a very slick environment. Please check it out!
Hmm I was in Ecuador for a month earlier this year (mostly working at a school in Quito but also traveling around in the Andes region to Riobamba) and I don't think I ever saw a power outage! I don't think most of the PCs at the school had UPSes and we never had trouble.
I *have* seen that in other countries -- Honduras has HORRIBLE power, even the big cities. But I haven't been there since 1994, so maybe things have improved.
After all that, it's a decent inventory management/accounting package, perhaps some decent MRP/ERP functionality.
GNU Enterprise and NOLA are going to hopefully get there at some point in the future.
GNUe is porting NOLA to its platform, and should be pretty slick. (I've been working on that some, but not much recently..... need to get back to it!)
I have a Nokia 8250 with AT&T, and I've only gotten one or two "spams" from AT&T itself. Doesn't seem like it's a huge problem at this point...
Granted that this is just gut instinct, but I really don't think that would fly. By any reasonable definition, a beta tester is NOT part of the company. In fact, going back to the original definitions, alpha testing is done inside the company and beta testing is done outside the company, once the product is deemed to be in somewhat stable shape internally.
Interesting.
The real question is whether or not a closed beta does indeed qualify the participant as an agent of the company.
I would guess that it does not. A closed beta recipient is (generally) not being paid by the company, has no permanent relationship, is installing the software himself, and is not using the company's hardware. IANAL.
However, UL might just be intending to apply the NDA to the non-GPL parts of the distro. Why would they care if someone took the gcc from their closed beta and redistributed it? The installer and config utilities are what they're concerned about (I think). Of course, if they have a heavily modified KDE, then they would NOT be able to keep their beta testers from redistributing that, including their modifications.
interesting. I was thinking of downloading knoppix to burn and give out to people. I have an HP internal 8x IDE burner (8100i or something like that). Got it in summer 2000. Would that be too old and unable to burn it?
(Lest someone just say "download it and try it", it would be nice to save the bandwidth if it won't work!)
How's this idea....
Simply require the user to start browsing at a web page, which of course lists "special deals" at local businesses. Make them click on at least one for more info. Then just let them surf for a while with no more interruptions.
Every time a new DHCP lease is assigned, make them go to this page and click an ad. Make the leases last maybe 2-4 hours.
I don't think that would annoy anyone too badly. The question is could you get enough interest from local businesses to fund the thing and feed yourself. Heck, if that would be profitable I might try to set something like that up here in Salem!
Now throw in plugins, so for instance the FS layer interprets JPEGs and adds extra attributes. Now you can read the colour depth of an image by doing "cat photo.jpg/colour_depth" or whatever. You can get the raw, uncompressed version of the file by doing "cp photo.jpg/raw > photo.raw".
Yikes! And this belongs in the kernel???
And even if it does belong there, wouldn't it be best in the general FS layer than in a specific FS? It could work across ALL filesystems...
Good points, but it's not nearly as bad as you make it sound. At least not anymore.
Linux Standard Base, people....
Actually, even though the Linux market is much smaller, there's less competition with other vendors there, so if you produce a quality program, you're pretty much guaranteed to get a good number of sales.
I can think of several applications that would probably sell well on Linux... lots of edutainment type programs (Oregon Trail, etc), a Print-Shop type program that would easily create greeting cards and banners and posters, a good personal finance program (which TheKompany apparently has in Kapital), a QuickBooks workalike that really is as easy as QB...
If ALL that software was available and hardware companies offered it as a preload option, seriously, why would more people not consider Linux?
These guys are years away from the really hard stuff like VBA, OLE
So I guess you haven't realized that OOo has a complete Basic macro language like VBA built in already...
No, it's not compatible with VBA (unfortunately) but it's definitely there.
I believe it also has OLE like functionality too, even on Linux. Not 100% sure of the specifics though.
(1) Some unexpected people are going to _need_ MS Office, point-blank. The popular financial package we use only 'exports' data to Excel. Not excel file formats; just Excel (via OLE or COM or whatever they're calling it this week.) Although it has a 'print to CSV' feature (don't ask), it saves the file with some silly Lotus-specific extension that OpenOffice doesn't believe is actually a CSV file. Although renaming should in principle be easy, the people who need to use this data are simply not up to the task of understanding (a) why they need to set their folder options to show all those funny little three-letter thingies at the end of the filename, and (b) why they can't just click the 'Excel' button like they used to. So the people that need to regularly manipulate the data in the financial database at a relatively low level need MSO.
:D I'm learning OOo macro programming and need some paying work....
Crimony, the solution to this is so easy it isn't funny. Write an OpenOffice Basic macro to look for the CSV file, rename it if necessary, get OOo to load it, maybe even automatically apply some custom formatting. You could even have it load all the CSVs at once if there are more than one.
Want to hire me to implement that?
hmm, granted I'm no power user at this point, but I'm pretty sure all that is possible. You might want to ask on the OpenOffice users list... seriously...
I agree that the documentation leaves a bit to be desired, but they ARE working on it.
For one thing, there's a new Software Development Kit with info on how to write plugins.
Here's a link to thew new developer documentation draft.
Or any office suite that supports macros?
OOo lets you put a Basic macro in a document and have it run when the doc is openned. Couldn't it, say, look around in your home directory for text files or OOo documents, load said documents, and "hide" interesting info from those documents in the current document?
Hmmmm