Yes, how all those Unix admins running 'proprietary' mission critical stuff where downtime costs millions have missed out on upgrading all the time! What have you got used to upgrading - your laptop? You're telling us it's a good idea to have to upgrade working systems running 1000s of users on a regular basis?
I'd say he's being pretty accurate, based on numerous companies I've worked with, whether in the UK or the Middle East. These things take time - even longer if you are talking about going for Red Hat replacing an existing Windows desktop installation.
Also, if you were a client of Sybase, 'asking' why they weren't keeping up could well not get you anywhere. Incidentally it's rather a good example of Linux not being ready for 'enterprise' apps. Regardless of whether it's technically capable or not, if the vendor doesn't support it, tough!
IBM's involvement means a great deal. If you're an IT manager following IT trends, you'll have read about Linux in the trade press and shrugged your shoulders with vague interest. When the IBM sales guy turns up to tell you his AIX box can run Linux, or the Sun guy is talking about Linux boxes running your webservers, then you start to take more notice.
Before, the evangelists were totally irrelevant and totally without credibility - in terms of who an IT Manager would take advice from.
Comparing a government department to a business is extremely valid. A national government has a number of 'business' tasks which are similar to those faced by any business and to which can be applied the same rules of profit and loss, value for money, effort vs results, etc, etc.
A government's customers are its citizens. It has to bill/tax them, just as a telco might, it has to trace them and hold information on them that cannot be lost under any circumstances, just as a bank might, it has to market itself to them, just as any company might to its own customers. Running these operations can be automated, made more efficient, etc, etc by using computers. Judging by the UK government's efforts, writing their own software to do so is absolutely the worst thing to do:)
The point is that a government's mandate should be to consider all options and make an informed, educated choice, based on all the criteria - which does not mean blindly mandating open source, just as it doesn't mean considering open source should be precluded.
I wouldn't call SA a developing country. And if a country is a developing country, they're even more unlikely to have the resources to write their own stuff, but will have dollars to buy in outside help, hardware and licences.
Equally, everyone prices according to the local market.
If you need to run your national steel monopoly on something like Oracle, you're going to need the budget to do it properly, or you don't do it all. Also, if you have very little budget, you're not going to have the resources to write, test and support that national telco billing app, let alone purchase the F15K you'd need to run it on.
Why not indeed. The point is that mandating this kind of thing can cause more problems than it solves. I'm happily selling my closed source software to the government and both parties are benefitting. Why should I then have to change my entire business model because of a rather shortsighted mandate from the government? Equally the govt could get spectacularly bad value - they rely on my software and are asking me to GPL it and change my entire business model. Fine - I'd do it, but at a price!
There are 100s of scenarios, but my original point is that blanket mandates for open source benefit noone.
Having lots of talent and maybe lots of software companies locally has nothing to do with adopting Open Source or not. It could even harm that local talent. Take a small software house creating an app for which release under, say, the GPL, makes no business sense. If the government mandates 'open source' and asks for all source code to be released to them, the result could be that the local company loses money and the local talent can go off and find another job. The app could have been well priced, full of funcionality, etc, etc. That wouldn't matter if a blanked 'open source only' mantra is what appears in government tenders.
What is truly shortsighted is using open-source software. It may seem to get the job done now, but if it ties you down, if you're stuck with it, then you're in trouble in the long run. Only a shortsighted view would measure open-source stuff as better.
Doesn't really matter to a business or government whether it's open or closed - it should offer the best deal and be an acceptable balance of factors. If you go 100% down, say, a Linux desktop route for the company's 30,000 PCs, you're still 'locked in' whether you have the source or not!
"They are very few times that an inhouse solution can't be used instead of an Oracle app or an Siebel solution."
If that's the case, why are they highly successful multi-billion dollar companies with hundreds of thousands of customers all over the world? Taking the example of the kinds of companies I often work with. Two sys admins, a couple of DBAs and an IT manager, working for a large manufacturing company. You're telling me they have the skills, the time and the inclination to knock up a quick competitor to Oracle financials over the weekend?
The market decides a lot of things. If Microsoft licences are expensive, fair enough. The logical follow on to that is not simply to mandate 'open source', but to look at all the alternative packages on the market.
I feel it's nonsense to declare any governemnt department or private institution an 'open source' zone, if the implication is that only open source solutions will be considered.
What happens when they need functionality that the open source world doesn't offer. I'm thinking of things like the scalability and availability features you get from the big Unix guys (and no, sticking 100 Dells together is not always the answer for big systems). What about when something like SAP, Oracle Apps, Siebel, etc, etc is required?
Support issues and costs are not instantly solved just because you can look at the source code. That is utterly irrelevant to most IT managers. The last thing govt IT workers I know want to be told is that they no longer need that support contract - they can just look at the code man'. That simply doesn't hack it in a large number of situations. If it does work, then use it, but it shouldn't be the sole policy.
No IT solution should be dismissed out of hand, whether closed or open.
I don't think there's anything particularly inevitable about the tide. Think about who buys Sun's machines - businesses, after solutions. SGI's machine is focussed on a very specific target market. It has few commercial apps, no reference sites in industry, etc. Customers needing that Oracle upgrade, storage consolidation and custom in-house developed app, across multiple global sites, are not going to install a 64 way Itanium Linux machine from SGI.
Yes, I must pop out and tell all those 1000s of customers all over the world to ditch all their Suns and IBMs and invest millions in some unproven chipset that doesn't even properly exist yet. At the same time they really ought to start using an operating system with minimal enterprise features, support, software and scalability.
However, if in a couple of years they need some cheap, expendable boxes that aren't mission critical - a farm of web server for example, I wouldn't be suprised if they'd choose Linux running on Opterons.
Hear hear! Overpriced compared to what? Intel addresses its own niche, but comes nowhere near what the Solaris/SPARCs, or other enterprise vendors offer in terms of reliability, scalability, etc.
Sadly the IT world doesn't entirely consist of little PCs running Linux and Apache. And no guys, sticking in hundreds of dual cpu Linux boxes does not miraculously address every IT need where a large number of cpus is required.
Have a quick sojourn in the world of real IT and learn something.
You missed out a receptionist, internal sys admin, marketing and public relations people, business development managers and salesman, at the very least.
There's a difference between having staff to actually create the product - you have to run the company and also actually go out and sell the product as well. The cost of running a small business in France is also extremely high, when you take into account the tax and social payments needed in that country.
Re:solution for one of the problems..
on
The New IT Crisis
·
· Score: 1
An interesting take on this, coming from my experience pushing Sun Rays to various customers. A large percentage of companies whose employees needed laptops, needed them so that they could work in different company premises. IE, salesman A would spend his time between different regional headquarters, often only plugging in his laptop when actually in an office. With a centralised thin client model, where he got his data from whichever office he was working in, he wouldn't need a laptop.
Clearlt, this doesn't apply to every case, but in a large number of situations, laptops are needed to solve a problem that a decent thin client solution would address in a much more manageable way, from whence came the savings.
Basically, like a lot of people, I am extremely lazy. I used to have all my CDs on the shelf and would listen to one, but then no bother to get up and change it.
Now, with 2000 plus songs on the computer, I work, wander round the house, whatever, with my entire collection playing. I have playlists for different artists, moods, etc and never need to get off my arse and change the CD.
The result is that I am more 'into' music than ever before, as listening to it is easy and makes me want to buy other albums from the artists I enjoy.
I can afford to buy CDs. As a student I couldn't afford to - so I pirated friends CDs onto tape. The result was that I listened to music I otherwise wouldn't have bought anyway. Now I can afford the CD, it's easier just to buy it than go through the hassle of taping/copying.
Why are people at Slashdot so down on Sun at the moment? They stuck with Unix when all the other vendors were going for NT, they make great hardware and have a great OS and certainly aren't losing market share to IBM or HP for God's sake. They're here to stay!
What's wrong with replacing CDE with Gnome?
If you want StarOffice for free, use OpenOffice? Companies weren't interested in StarOffice when it was free, now we have the best of both worlds - a free open source version and a professionally supportable product with a licence.
From the posts, it sounds like they didn't ignore OpenBSDm, they asked for an NDA to be signed. What's wrong with that?
Slashdot Mirror
Yes, how all those Unix admins running 'proprietary' mission critical stuff where downtime costs millions have missed out on upgrading all the time! What have you got used to upgrading - your laptop? You're telling us it's a good idea to have to upgrade working systems running 1000s of users on a regular basis?
Time to move to Sun, methinks. Stable, supported, reliable.
I'd say he's being pretty accurate, based on numerous companies I've worked with, whether in the UK or the Middle East. These things take time - even longer if you are talking about going for Red Hat replacing an existing Windows desktop installation.
Also, if you were a client of Sybase, 'asking' why they weren't keeping up could well not get you anywhere. Incidentally it's rather a good example of Linux not being ready for 'enterprise' apps. Regardless of whether it's technically capable or not, if the vendor doesn't support it, tough!
IBM's involvement means a great deal. If you're an IT manager following IT trends, you'll have read about Linux in the trade press and shrugged your shoulders with vague interest. When the IBM sales guy turns up to tell you his AIX box can run Linux, or the Sun guy is talking about Linux boxes running your webservers, then you start to take more notice.
Before, the evangelists were totally irrelevant and totally without credibility - in terms of who an IT Manager would take advice from.
Thanks. If price hadn't been an issue, would you have taken the Suns?
What's the app and why didn't you upgrade the Suns?
Comparing a government department to a business is extremely valid. A national government has a number of 'business' tasks which are similar to those faced by any business and to which can be applied the same rules of profit and loss, value for money, effort vs results, etc, etc.
:)
A government's customers are its citizens. It has to bill/tax them, just as a telco might, it has to trace them and hold information on them that cannot be lost under any circumstances, just as a bank might, it has to market itself to them, just as any company might to its own customers. Running these operations can be automated, made more efficient, etc, etc by using computers. Judging by the UK government's efforts, writing their own software to do so is absolutely the worst thing to do
The point is that a government's mandate should be to consider all options and make an informed, educated choice, based on all the criteria - which does not mean blindly mandating open source, just as it doesn't mean considering open source should be precluded.
Hmm, if you're using StarOffice 6.0, you really ought to be paying some kind of licence fee.
I wouldn't call SA a developing country. And if a country is a developing country, they're even more unlikely to have the resources to write their own stuff, but will have dollars to buy in outside help, hardware and licences.
Equally, everyone prices according to the local market.
If you need to run your national steel monopoly on something like Oracle, you're going to need the budget to do it properly, or you don't do it all. Also, if you have very little budget, you're not going to have the resources to write, test and support that national telco billing app, let alone purchase the F15K you'd need to run it on.
Why not indeed. The point is that mandating this kind of thing can cause more problems than it solves. I'm happily selling my closed source software to the government and both parties are benefitting. Why should I then have to change my entire business model because of a rather shortsighted mandate from the government? Equally the govt could get spectacularly bad value - they rely on my software and are asking me to GPL it and change my entire business model. Fine - I'd do it, but at a price!
There are 100s of scenarios, but my original point is that blanket mandates for open source benefit noone.
Having lots of talent and maybe lots of software companies locally has nothing to do with adopting Open Source or not. It could even harm that local talent. Take a small software house creating an app for which release under, say, the GPL, makes no business sense. If the government mandates 'open source' and asks for all source code to be released to them, the result could be that the local company loses money and the local talent can go off and find another job. The app could have been well priced, full of funcionality, etc, etc. That wouldn't matter if a blanked 'open source only' mantra is what appears in government tenders.
What is truly shortsighted is using open-source software. It may seem to get the job done now, but if it ties you down, if you're stuck with it, then you're in trouble in the long run. Only a shortsighted view would measure open-source stuff as better.
Doesn't really matter to a business or government whether it's open or closed - it should offer the best deal and be an acceptable balance of factors. If you go 100% down, say, a Linux desktop route for the company's 30,000 PCs, you're still 'locked in' whether you have the source or not!
"They are very few times that an inhouse solution can't be used instead of an Oracle app or an Siebel solution."
If that's the case, why are they highly successful multi-billion dollar companies with hundreds of thousands of customers all over the world? Taking the example of the kinds of companies I often work with. Two sys admins, a couple of DBAs and an IT manager, working for a large manufacturing company. You're telling me they have the skills, the time and the inclination to knock up a quick competitor to Oracle financials over the weekend?
The market decides a lot of things. If Microsoft licences are expensive, fair enough. The logical follow on to that is not simply to mandate 'open source', but to look at all the alternative packages on the market.
How much does SA spend on foreign made vehicles? Does that mean it's worth developing their own car manufacturing industry?
I feel it's nonsense to declare any governemnt department or private institution an 'open source' zone, if the implication is that only open source solutions will be considered.
What happens when they need functionality that the open source world doesn't offer. I'm thinking of things like the scalability and availability features you get from the big Unix guys (and no, sticking 100 Dells together is not always the answer for big systems). What about when something like SAP, Oracle Apps, Siebel, etc, etc is required?
Support issues and costs are not instantly solved just because you can look at the source code. That is utterly irrelevant to most IT managers. The last thing govt IT workers I know want to be told is that they no longer need that support contract - they can just look at the code man'. That simply doesn't hack it in a large number of situations. If it does work, then use it, but it shouldn't be the sole policy.
No IT solution should be dismissed out of hand, whether closed or open.
I don't think there's anything particularly inevitable about the tide. Think about who buys Sun's machines - businesses, after solutions. SGI's machine is focussed on a very specific target market. It has few commercial apps, no reference sites in industry, etc. Customers needing that Oracle upgrade, storage consolidation and custom in-house developed app, across multiple global sites, are not going to install a 64 way Itanium Linux machine from SGI.
StarOffice was cheap when the OpenOffice split occurred - less than $50 a licence.
How is it a dirty trick? The codebases were the same and OpenOffice caught up with StarOffice extremely quickly.
Why do you say that? Give us some examples please.
Yes, I must pop out and tell all those 1000s of customers all over the world to ditch all their Suns and IBMs and invest millions in some unproven chipset that doesn't even properly exist yet. At the same time they really ought to start using an operating system with minimal enterprise features, support, software and scalability.
However, if in a couple of years they need some cheap, expendable boxes that aren't mission critical - a farm of web server for example, I wouldn't be suprised if they'd choose Linux running on Opterons.
Hear hear! Overpriced compared to what? Intel addresses its own niche, but comes nowhere near what the Solaris/SPARCs, or other enterprise vendors offer in terms of reliability, scalability, etc.
Sadly the IT world doesn't entirely consist of little PCs running Linux and Apache. And no guys, sticking in hundreds of dual cpu Linux boxes does not miraculously address every IT need where a large number of cpus is required.
Have a quick sojourn in the world of real IT and learn something.
You missed out a receptionist, internal sys admin, marketing and public relations people, business development managers and salesman, at the very least.
There's a difference between having staff to actually create the product - you have to run the company and also actually go out and sell the product as well. The cost of running a small business in France is also extremely high, when you take into account the tax and social payments needed in that country.
An interesting take on this, coming from my experience pushing Sun Rays to various customers. A large percentage of companies whose employees needed laptops, needed them so that they could work in different company premises. IE, salesman A would spend his time between different regional headquarters, often only plugging in his laptop when actually in an office. With a centralised thin client model, where he got his data from whichever office he was working in, he wouldn't need a laptop.
Clearlt, this doesn't apply to every case, but in a large number of situations, laptops are needed to solve a problem that a decent thin client solution would address in a much more manageable way, from whence came the savings.
Basically, like a lot of people, I am extremely lazy. I used to have all my CDs on the shelf and would listen to one, but then no bother to get up and change it.
Now, with 2000 plus songs on the computer, I work, wander round the house, whatever, with my entire collection playing. I have playlists for different artists, moods, etc and never need to get off my arse and change the CD.
The result is that I am more 'into' music than ever before, as listening to it is easy and makes me want to buy other albums from the artists I enjoy.
I can afford to buy CDs. As a student I couldn't afford to - so I pirated friends CDs onto tape. The result was that I listened to music I otherwise wouldn't have bought anyway. Now I can afford the CD, it's easier just to buy it than go through the hassle of taping/copying.
Why are people at Slashdot so down on Sun at the moment? They stuck with Unix when all the other vendors were going for NT, they make great hardware and have a great OS and certainly aren't losing market share to IBM or HP for God's sake. They're here to stay!
What's wrong with replacing CDE with Gnome?
If you want StarOffice for free, use OpenOffice? Companies weren't interested in StarOffice when it was free, now we have the best of both worlds - a free open source version and a professionally supportable product with a licence.
From the posts, it sounds like they didn't ignore OpenBSDm, they asked for an NDA to be signed. What's wrong with that?