In the feedback in the/. article about that spec, someone pointed out that the 'no upgrade' was only part of one spec - the 'Easy PC' (which would imply to iPaqs and other enterprise- or consumer-friendly iMac type machines). The standard desktop PC spec actually has a requirement for easy hardware access. -- Assume that there are valid arguments against your position.
They probably should've prepped him (being a server guy, he probably isn't familiar with how the hardware licensing thingy works. Anyway...
The licensing system is supposed to be designed to allow components to change. It's based on the entire system, and will only complain when installed on a completely different system (and it will even allow being installed on several completely different systems before complaining). You could gradually swap out every component on your system, and it'll let you do that indefinitely.
I'd wait to see how well that works before assuming you're never going to be able to change anything... Microsoft isn't that stupid. What they're going after here are the same serial number being installed on 10 machines in a single week. -- Assume that there are valid arguments against your position.
- I'm not sure how you can separate politically from financially viable; the station cost estimates were continually exceeding the amount of money that was politically viable. NASA has become very very good at burning through money, which it continues to do to this day (hence $4 billion cost overrun on the ISS components).
- Standards are arguable; the Russian engineers are actually worried that the US components of station are much more vulnerable to explosive decompression if hit by space debris. NASA's stuff is usually more high-tech, but sometimes use of the more advanced technology is questionable (example: Shuttle's highly volatile solid rocket boosters)
- The Russians are providing elements that NASA just didn't have, such as reboost capability and safe return capability. In the last US design iteration before russians joined the project as a partner, the plan was to buy soviet Soyuz craft to use as crew return vehicles. True, most of the launches involved are American, but this is largely due to the collapse of the soviet economy. The Buran shuttle could easily be doing what the American shuttle is doing.
- Well, ten US astronauts have died (3 in the Apollo 1, 7 in Challenger accident) compared to four Soviet cosmonauts (1 in Soyuz 1, 3 in Soyuz 11). I don't have figures for who's had more launches, but since the Soviets were launching through the late 70s (while the US program was essentially halted) and during the post-Challenger investigations, I imagine the figures are comparable. And soviets would certainly have the a great lead in total man-hours in space.
In any case, arguments can certainly be for both sides of the 'who is better' argument, but Russian contributions to manned space exploration tend to be undeservedly disparaged. -- Assume that there are valid arguments against your position.
>> Just because you do something first doesn't make you better at it. The russians built a space station. The US went to the moon several times.
Well, not the argue the 'stepped out of bounds' on the international station, which I agree with, but a very good case could be made the the russians have had a much more successful space program.
The Russians
- have a better safety record
- have cheaper and more reliable expendable launch vehicles
- have had multiple generations of space stations
- have a highly reliable unmanned supply system (the Progress spacecraft)
- have their own shuttle (the Buran shuttle, which flew once before being mothballed due to lack of funds) and a heavy-lift system (Energiya) which could be used to loft the shuttle or any other heavy components (a station the mass of Mir could be lifted in one shot)
The americans, on the other hand, threw out pretty much everything they had a put all their eggs in the far-too-ambitious, far-too-expensive Space Shuttle. And after they did that, they threw billions in the far-too-ambitious, far-too-expensive Space Station program. And it wasn't until the Russians stepped in (close to 10 years and 10 billion dollars after the US Space Station program began) that the US station became viable.
After the success of the Apollo program, the US apparently forgot that the best way to get to space is to start simple and work your way up. Their initial designs for the space station had freakin' hangars to repair satellites in. Looked cool, sure, but wholely impractical.
The Russians have demonstrated a much better manned space program. If it wasn't for their whole economy collapsing, they'd have a 2nd-generation permanent station, a cheap and reliable launch and resupply system, a shuttle for when the capabilities of shuttles were needed, and a heavy-launch vehicle. The States would probably still be squandering billions on a station with no crew escape capabilities, no self-propulsion abilities, an enormously expensive resupply system (the shuttle) and increasingly reduced science capabilities. -- Assume that there are valid arguments against your position.
He's trained on the Russian side (ie. he was trained for Mir, and then given updates for the Russian side of the ISS). He's not trained on the American side. -- Assume that there are valid arguments against your position.
All the other international partners have been opposed to Tito in lockstep with NASA since the beginning. The ESA, in particular, is opposed to it... There's an german astronaut currently in training; if this situation is amicably resolved, it will likely involve ESA paying for this guy to go up instead of Tito.
No one is challenging that Tito should be able to go up at some point. The issues, however, are that (a) he needs 6-8 weeks of training on the non-Russian side of the station, which he can't get before the flight, and (b) he doesn't have any sort of insurance for this sort of thing (and no one really knows what's required there).
However, there's no arguing that Russia really needs the money... the RSA's apparently been allocated by the Russian government about $150 million, so the $20 million that Tito is paying is not inconsequencial (like it would be for NASA).
Tito himself says he's unavailable for the next flight (in October) when the training and insurance issues could be resolved. How someone who's got $20 million to throw around can't make himself available any time he wants, I don't get, but anywayz... -- Assume that there are valid arguments against your position.
Well, that's remarkably easy if you're a UNIX geek that knows all those commands. Which would probably only include a subset of UNIX geeks.
10 years and you still gotta know commands like 'vnc' and 'fg'? That's why Linux isn't taking over the average desktop any time soon.
My Mom could use the Windows XP terminal services. Any computer-literate person could use Windows terminal services in 2 seconds without having to think of looking up MANs and HOWTOs and newsgroup postings... -- Assume that there are valid arguments against your position.
it would just have to check submitted songs to make sure they're not on the disapproved list.
So if you said "I'd like to submit the song 'Happy Fun Band - Dumb Song.mp3' to be put on the approval list" Napster would approve it. And if you said "I'd like to submit the song 'aMetallica - aDisappear.mp3'" Napster would tell you to piss off.
It would certainly be easier for them to filter individual requests than to try to run searches on every song every time a user logs in. And once they get the subscription service going, it's easy for them to add all the BMG tunes.
-- Assume that there are valid arguments against your position.
MLB would distribute an even cut of the money to each team, even though some teams like the O's have stronger radio audiences than others
One could argue that that's a good thing, as teams with large radio audiences would seem to be teams with large local support, and thus far more money than smaller teams anyway. (ie. New York) -- Assume that there are valid arguments against your position.
> By default though, non-admin user's do have the rights to shutdown (on professional, at least)
It's probably assumed on workstations you've got access to the physical machine... which means you've got access to power off no matter what. So you might as well let them close the system properly.
On Windows Server systems, only Admins have Shutdown priviledges... -- Assume that there are valid arguments against your position.
Um, no it doesn't mean that at all. Why would listing the programs running imply control over them?
It just means that if Dad walks away from the machine with 10 programs running, and Jr walks up to use the machine, he can see Jr has 10 programs running and know that Counterstrike probably isn't going to have a good frame rate at 1280 resolution... Also helpful, Mom can walk up and see if she has new e-mail without logging into the machine.
As a regular user of Terminal Services, I can tell you that MS handles multiuser OS functions beautifully. You can even run a session at work (these are, of course, fully GUI sessions which run acceptably even over a modem), Disconnect (which leaves the session running), and log into the same session from a completely different machine with everything still running just the way you left it...
-- Assume that there are valid arguments against your position.
In my experience, while the Athlon chip itself is great, its motherboard chipsets just don't seem to be up par with the chipsets for Intel systems in terms of driver stability and compatibility. -- Assume that there are valid arguments against your position.
...is a good one. But it's more of a tribute to Star Trek fan pages; it wasn't ever intended to appear genuine. Same way the movie was kind of a tribute to Trek fans. -- Assume that there are valid arguments against your position.
> Both are very good products/projects, so why standarize on either?
Well, if you're writing, say, a work processing app, which environment do you design it for? Now, how well will it work with some else's graphic app, designed for a different environment? And will both of them work with the spreadsheet someone else wrote?
Will they share the same clipboard? Use the same printer drivers and control panels? Can you embed spreadsheets and pictures into the word processor, and double-click on them to edit?
And even if all that could be made to work (and I understand that efforts along those lines are being made in KDE/Gnome compatibility) they're still going to have inconsistent look & feels.
>> I just got a/. article rejected today where I
>> explained all this plus gave links to all the
>> free MPEG-4 implementations and the Sorenson >> MPEG-4 press release.
Er, any chance you could put those links up anyway?;)
> 1) The "hacker-kiddies" ain't gonna sue for
> trademark infringement, and more importantly:
Yes, but the original DivX people could. I dunno if they would, seeing as the whole things gone to pot, but Circuit City (or whoever) probably still owns the DivX trademark. Of course, on DivXnetworks their website it says 'DivX(tm)' so maybe they bought the name or something.
> 2) Without the name-recognition of the geek
> set, where are you gonna generate the "buzz"
> required for a successful financing?
I think the general population is more likely to recognize DivX as the lousy Circuit City scheme than the decoding scheme. So now whenever they try to sell it, they have to explain what it is.
(Of course, most of the population probably hasn't heard of either, so it probably doesn't matter.)
I think my point is that the whole name was dumbass to begin with. It like forming a bus company and calling it, like, "TWA".
SOAP is basically a series of XML specs saying how you access objects/functions/arguments etc on the server machine. So you pass an XML document to a SOAP server, it processes it, and returns you an XML document with the results.
.NET's Web Services is simply an easy way of processing this, as is Sun's similar project, as is soap. You just write your code with your objects, functions, etc and let.NET handle the XML bits.
You could, however, write a SOAP object just using, for example, a Perl script which parses the incoming XML manually (it comes through HTTP POST) and then spits out the XML response just like any other web page.
What's interesting is a whole lot of this.NET initiative (the Common Lanuage Runtime, the C-sharp programming language) came about after Sun sued Microsoft out of Java.
They claim they'll have a complicated algorithm system that'll track the changes and try to decide what constitutes "reasonable" changes.
So it'll let you do complete swap outs (essentially get a totally new pc) a couple of times; but if you just gradually install one component after another, it'll just track those changes and let you do that, like, infinitely.
It's basically not going to start squeeling until it's being used 3 times a day on totally different machines.
In theory, anyway. (And the whole privacy issue here is another can o' worms.)
Windows Update tends only to include client-related security patches; server-related patches are listed on a separate Corporate Windows Update page and aren't automatically sorted by what's installed on your system. Part of the problem is people think Windows Update covers everything when it doesn't really cover anything server-related (until a service pack is released). For the list of server-related patches, go to http://corporate.windowsupdate.microsoft.com/
You gotta download them from this site, run them, then (usually) reboot. That's the problem #1 with Microsoft patches: instead of stopping the service (say IIS), installing the files, and restarting the service (which would work fine for 95% of patches) they insist on a reboot. Which is nuts for a server - expecially when new patches are released every week or two.
(Tip: You can install all of them at once (in a batch file or just running them individually) then reboot. Works fine.)
Seems the number #1 problem you got here is Access on the Macintosh is a dead end which hasn't been updated for years. So it's basically not an option if yours is a mixed shop.
Otherwise, using Access with myODBC works fairly well for what Access is good at: a bunch of fairly simple forms and reports. Access' wizards and such make simple data input forms a snap, and it's reporting engine is very easy to use... certainly better than any of the web one out there if you need special formatting and the like.
The problem with Access is you lose a lot of control; Access tends to do lots of things for you, even if you don't especially want it to. I work in a Microsoft shop, and I tend to use VB instead of Access when a client app is desired... it's data access components are nearly as good as Access, but give you a lot more control.
But Web apps are definitely a better way to go. No worries about client-side drivers and everything... no worries about configuring the myODBC drivers and the ODBC settings and all... and no cross-platform concerns.
If this app is supposed to work on both Macs and PCs, that's your killer argument: cross-platform development is a bitch, and Access on the Mac isn't supported anymore.
If it's only supposed to work on PCs, point out the configuration and compatibility issues involved in installing client-side apps to multiple PCs. Unless it's a fairly simple app with heavy reporting needs, in which case it might be worth looking into Access.
If it's only supposed to work on one PC, take a look a Access' capabilities. It might save you days of work.
Small correction: Access + SQL Server doesn't use Jet. Jet is the mechanism for using native Access file format (.MDB) - which Microsoft is actually phasing out.
Access is actually going through a transition with Access 2000 to a proper client-server system:
The old "Access database" format uses the DAO object model to access.MDB files (Jet engine) natively and everything else through ODBC.
The newer "Access project" format uses the ADO object model (which is much nicer than DAO) to access many different data formats (including SQL Server) natively, as well as ODBC. Access 2000 comes with a mini version of SQL Server called the Microsoft Data Engine (MSDE) that can be distributed for free. Access Projects are meant to be written for the MSDE, and can be scaled up to SQL Server without any new code at all.
As actually using it: It depends totally on what's meant to be accomplished. If it's just a a couple of simple forms and reports, they can be whipped out in an afternoon with Access. If it gets more complicated, all the stuff Access does for you automatically (updates, etc.) start getting in the way.
Well, all you have to do is e-mail your favorite web sites (you can start with Slashdot), tell them to stop using the banner ads, and suggest how they make up the traffic/revenue they got from them.
> The short story is the money isn't in software. > It probably never was.
Um, yes, the money is. IBM actually makes all its money of software (and software consulting). There's very little profit in their hardware devision - and a small portion of their revenue comes from their Linux work, although of course this is growing. And Apple's business fluctuates far too much to be an example of successful business.
And most of Microsoft's profits come from selling Microsoft Office to business costumers. The money is made selling to businesses (who *have* to have legitimate copies of software).
What might be accurate is to say the money in the home market isn't in software. That might be true... except for, like, games.
But try telling Oracle, at $100,000+ a pop for their database, that the money isn't in software.;)
Slashdot Mirror
In the feedback in the /. article about that spec, someone pointed out that the 'no upgrade' was only part of one spec - the 'Easy PC' (which would imply to iPaqs and other enterprise- or consumer-friendly iMac type machines). The standard desktop PC spec actually has a requirement for easy hardware access.
--
Assume that there are valid arguments against your position.
They probably should've prepped him (being a server guy, he probably isn't familiar with how the hardware licensing thingy works. Anyway...
The licensing system is supposed to be designed to allow components to change. It's based on the entire system, and will only complain when installed on a completely different system (and it will even allow being installed on several completely different systems before complaining). You could gradually swap out every component on your system, and it'll let you do that indefinitely.
I'd wait to see how well that works before assuming you're never going to be able to change anything... Microsoft isn't that stupid. What they're going after here are the same serial number being installed on 10 machines in a single week.
--
Assume that there are valid arguments against your position.
- I'm not sure how you can separate politically from financially viable; the station cost estimates were continually exceeding the amount of money that was politically viable. NASA has become very very good at burning through money, which it continues to do to this day (hence $4 billion cost overrun on the ISS components).
- Standards are arguable; the Russian engineers are actually worried that the US components of station are much more vulnerable to explosive decompression if hit by space debris. NASA's stuff is usually more high-tech, but sometimes use of the more advanced technology is questionable (example: Shuttle's highly volatile solid rocket boosters)
- The Russians are providing elements that NASA just didn't have, such as reboost capability and safe return capability. In the last US design iteration before russians joined the project as a partner, the plan was to buy soviet Soyuz craft to use as crew return vehicles. True, most of the launches involved are American, but this is largely due to the collapse of the soviet economy. The Buran shuttle could easily be doing what the American shuttle is doing.
- Well, ten US astronauts have died (3 in the Apollo 1, 7 in Challenger accident) compared to four Soviet cosmonauts (1 in Soyuz 1, 3 in Soyuz 11). I don't have figures for who's had more launches, but since the Soviets were launching through the late 70s (while the US program was essentially halted) and during the post-Challenger investigations, I imagine the figures are comparable. And soviets would certainly have the a great lead in total man-hours in space.
In any case, arguments can certainly be for both sides of the 'who is better' argument, but Russian contributions to manned space exploration tend to be undeservedly disparaged.
--
Assume that there are valid arguments against your position.
>> Just because you do something first doesn't make you better at it. The russians built a space station. The US went to the moon several times.
Well, not the argue the 'stepped out of bounds' on the international station, which I agree with, but a very good case could be made the the russians have had a much more successful space program.
The Russians
- have a better safety record
- have cheaper and more reliable expendable launch vehicles
- have had multiple generations of space stations
- have a highly reliable unmanned supply system (the Progress spacecraft)
- have their own shuttle (the Buran shuttle, which flew once before being mothballed due to lack of funds) and a heavy-lift system (Energiya) which could be used to loft the shuttle or any other heavy components (a station the mass of Mir could be lifted in one shot)
The americans, on the other hand, threw out pretty much everything they had a put all their eggs in the far-too-ambitious, far-too-expensive Space Shuttle. And after they did that, they threw billions in the far-too-ambitious, far-too-expensive Space Station program. And it wasn't until the Russians stepped in (close to 10 years and 10 billion dollars after the US Space Station program began) that the US station became viable.
After the success of the Apollo program, the US apparently forgot that the best way to get to space is to start simple and work your way up. Their initial designs for the space station had freakin' hangars to repair satellites in. Looked cool, sure, but wholely impractical.
The Russians have demonstrated a much better manned space program. If it wasn't for their whole economy collapsing, they'd have a 2nd-generation permanent station, a cheap and reliable launch and resupply system, a shuttle for when the capabilities of shuttles were needed, and a heavy-launch vehicle. The States would probably still be squandering billions on a station with no crew escape capabilities, no self-propulsion abilities, an enormously expensive resupply system (the shuttle) and increasingly reduced science capabilities.
--
Assume that there are valid arguments against your position.
>Tito is already trained.
He's trained on the Russian side (ie. he was trained for Mir, and then given updates for the Russian side of the ISS). He's not trained on the American side.
--
Assume that there are valid arguments against your position.
All the other international partners have been opposed to Tito in lockstep with NASA since the beginning. The ESA, in particular, is opposed to it... There's an german astronaut currently in training; if this situation is amicably resolved, it will likely involve ESA paying for this guy to go up instead of Tito.
No one is challenging that Tito should be able to go up at some point. The issues, however, are that (a) he needs 6-8 weeks of training on the non-Russian side of the station, which he can't get before the flight, and (b) he doesn't have any sort of insurance for this sort of thing (and no one really knows what's required there).
However, there's no arguing that Russia really needs the money... the RSA's apparently been allocated by the Russian government about $150 million, so the $20 million that Tito is paying is not inconsequencial (like it would be for NASA).
Tito himself says he's unavailable for the next flight (in October) when the training and insurance issues could be resolved. How someone who's got $20 million to throw around can't make himself available any time he wants, I don't get, but anywayz...
--
Assume that there are valid arguments against your position.
Well, that's remarkably easy if you're a UNIX geek that knows all those commands. Which would probably only include a subset of UNIX geeks.
10 years and you still gotta know commands like 'vnc' and 'fg'? That's why Linux isn't taking over the average desktop any time soon.
My Mom could use the Windows XP terminal services. Any computer-literate person could use Windows terminal services in 2 seconds without having to think of looking up MANs and HOWTOs and newsgroup postings...
--
Assume that there are valid arguments against your position.
it would just have to check submitted songs to make sure they're not on the disapproved list.
So if you said "I'd like to submit the song 'Happy Fun Band - Dumb Song.mp3' to be put on the approval list" Napster would approve it. And if you said "I'd like to submit the song 'aMetallica - aDisappear.mp3'" Napster would tell you to piss off.
It would certainly be easier for them to filter individual requests than to try to run searches on every song every time a user logs in. And once they get the subscription service going, it's easy for them to add all the BMG tunes.
--
Assume that there are valid arguments against your position.
One could argue that that's a good thing, as teams with large radio audiences would seem to be teams with large local support, and thus far more money than smaller teams anyway. (ie. New York)
--
Assume that there are valid arguments against your position.
> By default though, non-admin user's do have the rights to shutdown (on professional, at least)
It's probably assumed on workstations you've got access to the physical machine... which means you've got access to power off no matter what. So you might as well let them close the system properly.
On Windows Server systems, only Admins have Shutdown priviledges...
--
Assume that there are valid arguments against your position.
Um, no it doesn't mean that at all. Why would listing the programs running imply control over them?
It just means that if Dad walks away from the machine with 10 programs running, and Jr walks up to use the machine, he can see Jr has 10 programs running and know that Counterstrike probably isn't going to have a good frame rate at 1280 resolution... Also helpful, Mom can walk up and see if she has new e-mail without logging into the machine.
As a regular user of Terminal Services, I can tell you that MS handles multiuser OS functions beautifully. You can even run a session at work (these are, of course, fully GUI sessions which run acceptably even over a modem), Disconnect (which leaves the session running), and log into the same session from a completely different machine with everything still running just the way you left it...
--
Assume that there are valid arguments against your position.
In my experience, while the Athlon chip itself is great, its motherboard chipsets just don't seem to be up par with the chipsets for Intel systems in terms of driver stability and compatibility.
--
Assume that there are valid arguments against your position.
...is a good one. But it's more of a tribute to Star Trek fan pages; it wasn't ever intended to appear genuine. Same way the movie was kind of a tribute to Trek fans.
--
Assume that there are valid arguments against your position.
> Both are very good products/projects, so why standarize on either?
Well, if you're writing, say, a work processing app, which environment do you design it for? Now, how well will it work with some else's graphic app, designed for a different environment? And will both of them work with the spreadsheet someone else wrote?
Will they share the same clipboard? Use the same printer drivers and control panels? Can you embed spreadsheets and pictures into the word processor, and double-click on them to edit?
And even if all that could be made to work (and I understand that efforts along those lines are being made in KDE/Gnome compatibility) they're still going to have inconsistent look & feels.
>> I just got a /. article rejected today where I
;)
>> explained all this plus gave links to all the
>> free MPEG-4 implementations and the Sorenson >> MPEG-4 press release.
Er, any chance you could put those links up anyway?
> 1) The "hacker-kiddies" ain't gonna sue for
> trademark infringement, and more importantly:
Yes, but the original DivX people could. I dunno if they would, seeing as the whole things gone to pot, but Circuit City (or whoever) probably still owns the DivX trademark. Of course, on DivXnetworks their website it says 'DivX(tm)' so maybe they bought the name or something.
> 2) Without the name-recognition of the geek
> set, where are you gonna generate the "buzz"
> required for a successful financing?
I think the general population is more likely to recognize DivX as the lousy Circuit City scheme than the decoding scheme. So now whenever they try to sell it, they have to explain what it is.
(Of course, most of the population probably hasn't heard of either, so it probably doesn't matter.)
I think my point is that the whole name was dumbass to begin with. It like forming a bus company and calling it, like, "TWA".
"DivX" is the reviled and defunct pay-per-play product.
;)" is an Mpeg4 encoding thingy used mainly to encode porn and pirated movies.
"DivX
See, it was given the name by hacker kiddies as a jab against the DivX format. Funny, huh?
How the hell they can form a company, get $5.6 million in venture funding, and still keep the hacker kiddie name I just don't get.
SOAP is basically a series of XML specs saying how you access objects/functions/arguments etc on the server machine. So you pass an XML document to a SOAP server, it processes it, and returns you an XML document with the results.
.NET handle the XML bits.
.NET's Web Services is simply an easy way of processing this, as is Sun's similar project, as is soap. You just write your code with your objects, functions, etc and let
You could, however, write a SOAP object just using, for example, a Perl script which parses the incoming XML manually (it comes through HTTP POST) and then spits out the XML response just like any other web page.
What's interesting is a whole lot of this .NET initiative (the Common Lanuage Runtime, the C-sharp programming language) came about after Sun sued Microsoft out of Java.
They claim they'll have a complicated algorithm system that'll track the changes and try to decide what constitutes "reasonable" changes.
So it'll let you do complete swap outs (essentially get a totally new pc) a couple of times; but if you just gradually install one component after another, it'll just track those changes and let you do that, like, infinitely.
It's basically not going to start squeeling until it's being used 3 times a day on totally different machines.
In theory, anyway. (And the whole privacy issue here is another can o' worms.)
Windows Update tends only to include client-related security patches; server-related patches are listed on a separate Corporate Windows Update page and aren't automatically sorted by what's installed on your system. Part of the problem is people think Windows Update covers everything when it doesn't really cover anything server-related (until a service pack is released). For the list of server-related patches, go to http://corporate.windowsupdate.microsoft.com/
You gotta download them from this site, run them, then (usually) reboot. That's the problem #1 with Microsoft patches: instead of stopping the service (say IIS), installing the files, and restarting the service (which would work fine for 95% of patches) they insist on a reboot. Which is nuts for a server - expecially when new patches are released every week or two.
(Tip: You can install all of them at once (in a batch file or just running them individually) then reboot. Works fine.)
Seems the number #1 problem you got here is Access on the Macintosh is a dead end which hasn't been updated for years. So it's basically not an option if yours is a mixed shop.
Otherwise, using Access with myODBC works fairly well for what Access is good at: a bunch of fairly simple forms and reports. Access' wizards and such make simple data input forms a snap, and it's reporting engine is very easy to use... certainly better than any of the web one out there if you need special formatting and the like.
The problem with Access is you lose a lot of control; Access tends to do lots of things for you, even if you don't especially want it to. I work in a Microsoft shop, and I tend to use VB instead of Access when a client app is desired... it's data access components are nearly as good as Access, but give you a lot more control.
But Web apps are definitely a better way to go. No worries about client-side drivers and everything... no worries about configuring the myODBC drivers and the ODBC settings and all... and no cross-platform concerns.
If this app is supposed to work on both Macs and PCs, that's your killer argument: cross-platform development is a bitch, and Access on the Mac isn't supported anymore.
If it's only supposed to work on PCs, point out the configuration and compatibility issues involved in installing client-side apps to multiple PCs. Unless it's a fairly simple app with heavy reporting needs, in which case it might be worth looking into Access.
If it's only supposed to work on one PC, take a look a Access' capabilities. It might save you days of work.
Small correction: Access + SQL Server doesn't use Jet. Jet is the mechanism for using native Access file format (.MDB) - which Microsoft is actually phasing out.
.MDB files (Jet engine) natively and everything else through ODBC.
Access is actually going through a transition with Access 2000 to a proper client-server system:
The old "Access database" format uses the DAO object model to access
The newer "Access project" format uses the ADO object model (which is much nicer than DAO) to access many different data formats (including SQL Server) natively, as well as ODBC. Access 2000 comes with a mini version of SQL Server called the Microsoft Data Engine (MSDE) that can be distributed for free. Access Projects are meant to be written for the MSDE, and can be scaled up to SQL Server without any new code at all.
As actually using it: It depends totally on what's meant to be accomplished. If it's just a a couple of simple forms and reports, they can be whipped out in an afternoon with Access. If it gets more complicated, all the stuff Access does for you automatically (updates, etc.) start getting in the way.
Well, all you have to do is e-mail your favorite web sites (you can start with Slashdot), tell them to stop using the banner ads, and suggest how they make up the traffic/revenue they got from them.
Simple.
> The short story is the money isn't in software. > It probably never was.
;)
Um, yes, the money is. IBM actually makes all its money of software (and software consulting). There's very little profit in their hardware devision - and a small portion of their revenue comes from their Linux work, although of course this is growing. And Apple's business fluctuates far too much to be an example of successful business.
And most of Microsoft's profits come from selling Microsoft Office to business costumers. The money is made selling to businesses (who *have* to have legitimate copies of software).
What might be accurate is to say the money in the home market isn't in software. That might be true... except for, like, games.
But try telling Oracle, at $100,000+ a pop for their database, that the money isn't in software.