COBOL Trends
in [Cobol]
|
From: Paul on 10 Sep 2009 22:16 I absolutely hate it when Pete's prognostications come true, but COBOL is becoming near impossible to get and use on mainstream (i.e. Windows and Linux) platforms. A lot of my (personal) software is written in IBM VisualAge COBOL, and for very good reasons. IBM gave Partners a REALLY big break on the price, making development feasible, and it doesn't have any runtime cost. Worked great. But it does not run on any modern version of Windows, at least not the GUI developer. I ported the software to VMS COBOL, which I love. I was thinking this would give me another 5 to 10 years. But the only machines my customers can buy that it runs on are Itanium based servers from HP. My customers are screaming for Windows and Macs refusing to buy Itaniums. I have mainframe software, but IBM no longer sells a mainframe for the "entry level" market. My customers on Flex and MP3K's are running out of time and have no place to go - but they sure know where they WANT to go - Windows, Macs, or Linux. I have AS/400 software, but the darn platform is going into oblivion - take it's new name for instance "i". Of course, there is still AIX COBOL, but that is being pushed hard and heavy by Linux. Then you go talk to the existing COBOL vendors today. Microfocus wants $4K for the NetExpress Developer - and $22K per server. Perhaps that is *per core* - I don't know, I was so shocked by their price increase I didn't bother asking. And they have sucked up both AcuCOBOL and Liant/RM. No options. Veryant COBOL sounds interesting, but they want a great deal of money for their development platform, and their runtimes are "half the cost" of MicroFocus. $11 per core is not cheap. :/ These Alchemy guys that took over NetCOBOL from Fujitsu are arrogant beyond belief. After just a short interaction with them, I think I would bite myself in the small of my back before I would send them money. All these vendors seem to have hit upon the idea of making COBOL "exclusive." Or something. Charge high for it and people will flock to buy it. Good luck with that. The sole exception that *I* have found is PerCOBOL. They seem to be excited about what they are doing, and lo and behold, they can and will talk intelligently with a person! Even discuss the benefits of competing products. I've not yet purchased their development set, but we may come to mutually happy agreement. I like their compiler, and it is a *easy* entryway to Java for a COBOL programs. (You guys might want to check it out if you are doing projects. I'm not affiliated with them in any way, but just the fact they will talk to you and try to help you understand their products is a very big plus in my book.) But back to Pete's ideas - darn if one of the best options for porting my COBOL code to something a bit more modern hasn't turned out to be Microsoft C#. :) The object orientation is not the draw for me, it is the really good support (albeit only under Windows) and the low cost of entry. No runtimes, and a minimum of $199 for the standard version of Visual Studio. $600 for the "Professional" version, which adds a few nice touches. So to me, the COBOL world, in which vendors where accommodating, welcoming, developer friendly, and interested in their customers has virtually disappeared, with the possible exception of the new PerCOBOL people. The future looks bleak folks- I sure hate this. Anyone else getting a different picture of the COBOL world? I would sure love to hear about it if you are. -Paul (Passing virtual six-pack over to Pete now... :)
From: Pete Dashwood on 11 Sep 2009 01:56 Paul wrote: > I absolutely hate it when Pete's prognostications come true, but COBOL > is becoming near impossible to get and use on mainstream (i.e. Windows > and Linux) platforms. I don't always like it either, Paul... :-) Nevertheless, there is no denying reality. > > A lot of my (personal) software is written in IBM VisualAge COBOL, and > for very good reasons. IBM gave Partners a REALLY big break on the > price, making development feasible, and it doesn't have any runtime > cost. Worked great. But it does not run on any modern version of > Windows, at least not the GUI developer. > > I ported the software to VMS COBOL, which I love. I was thinking this > would give me another 5 to 10 years. But the only machines my > customers can buy that it runs on are Itanium based servers from HP. > My customers are screaming for Windows and Macs refusing to buy > Itaniums. > I have mainframe software, but IBM no longer sells a mainframe for the > "entry level" market. My customers on Flex and MP3K's are running out > of time and have no place to go - but they sure know where they WANT > to go - Windows, Macs, or Linux. > It's a lot like the old VHS vs Betamax... the best product doesn't always win. As long as the product can do what most people want, and given enough marketing clout, there is a fighting chance it will succeed. Often, once the field has been won, the very success of the product causes it to get better. We have seen this with Windows. Windows 7 is a far cry from Win 3.1. (And no matter what else you may think about it, at least it isn't Vista... :-)) OK, that's not fair... Vista did finally improve and today's Vista is much bettere than the early releases.) Given that you are looking at migrating COBOL to Windows, maybe .NET long term, I think I can help. There are some options that I don't want to discuss publicly, that might be of interest to you. > I have AS/400 software, but the darn platform is going into oblivion - > take it's new name for instance "i". Of course, there is still AIX > COBOL, but that is being pushed hard and heavy by Linux. > > Then you go talk to the existing COBOL vendors today. Microfocus > wants $4K for the NetExpress Developer - and $22K per server. Perhaps > that is *per core* - I don't know, I was so shocked by their price > increase I didn't bother asking. And they have sucked up both AcuCOBOL > and Liant/RM. No options. Yes, it is a rough world out there and the demeanour of vendors seems to be changing as the screw turns. It may well be that for the market Michael described as being prime Micro Focus target, these prices are reasonable and even good value, but for a small business they are simply not viable. (I know from my own experience...) I remember working on contract for Rank Xerox in Germany many years ago. For around 25 years, they held the copyright to the photo copying process. Only game in town... they were rolling in it and their "salesmen" were really just "order takers". Then the copyright expired and suddenly there were other players in the market. They posted a loss of 25 million DM. First time in decades. I remember a Manager saying: "It's not important. To a company like us that is no big deal..." I asked him how many photo copies you need to make to get 25 million DM. He went very quiet. A small business has to ask the same kind of question. Before I spend 22K for my Net Express server, how long will it take to pay for itself? What other options do I have that could get me a similar result for less? The notebook I am writing this on gets replaced every 3 years (I'm acquiring quite a few of them :-)) But before I go and buy a new one, I think about how it will pay for itself. Writing software and articles can generate revenue but there is no guarantee that it will. Fortunately, the price of a notebook is not so high that I can't reasonably take a punt on it. This one has paid for itself many times over, so I guess it owes me nothing... But I never just go and buy one when 3 years is up. It has to justify its purchase or I make do with the old one a bit longer... > > Veryant COBOL sounds interesting, but they want a great deal of money > for their development platform, and their runtimes are "half the cost" > of MicroFocus. $11 per core is not cheap. :/ It is a market place. If people will pay these prices, then vendors will charge them. They are relying on the fact that most people just want to keep on doing what they've always done. Change requires effort. And time. Most of the existing COBOL market is simply bewildered by new technology, can't understand where all the COBOL went, and don't realise that as the COBOL world dries up, it is a smaller pond and vendors who want to fish in it HAVE to charge more. On the other hand they see the Windows pond (for example) becoming an Ocean where Microsoft can afford to GIVE development tools away for FREE! (I don't know how much longer this can last...I'm thinking they will start charging for these products once they are sure that Java is not a threat... That's another reason why I'm urging people to move now, rather than later. Get it while its free... :-)) I liked the attitude and enthusiasm of Veryant when we discussed their isCOBOL product, even though their tech guy couldn't understand what I said... :-) (antipodean accents don't render well on a poor connection...) > > These Alchemy guys that took over NetCOBOL from Fujitsu are arrogant > beyond belief. After just a short interaction with them, I think I > would bite myself in the small of my back before I would send them > money. That has, sadly, been my own experience with them too. It is a shame. The products are excellent and there is no runtime fee. I continue using the products but I am under no illusion as to what would happen if I ever needed maintenance. Fortunately, I don't. I have become quite adept at sorting out their software and really don't need their support. I'm using version 6 of NetCOBOL, which is the last one I licensed. I have helped people with problems in later versions and it seems to me there is very little difference between V6 and V9. I can understand people not being happy to pay $800 a year for something that doesn't change significantly. As for support, it was abysmal anyway (hopefully Alchemy will have addressed this; Fujitsu USA was just hopeless after they fired the people who knew anything.) and you have more chance of getting free answers from this forum than you ever had from the official sources. > > All these vendors seem to have hit upon the idea of making COBOL > "exclusive." Or something. Charge high for it and people will flock to > buy it. Good luck with that. They really don't have much choice. As the pond evaporates the price of water has to go up... The answer (for me at least) is move to a different pond... > > The sole exception that *I* have found is PerCOBOL. They seem to be > excited about what they are doing, and lo and behold, they can and > will talk intelligently with a person! Even discuss the benefits of > competing products. I've not yet purchased their development set, but > we may come to mutually happy agreement. I like their compiler, and it > is a *easy* entryway to Java for a COBOL programs. (You guys might > want to check it out if you are doing projects. I'm not affiliated > with them in any way, but just the fact they will talk to you and try > to help you understand their products is a very big plus in my book.) > > But back to Pete's ideas - darn if one of the best options for porting > my COBOL code to something a bit more modern hasn't turned out to be > Microsoft C#. :) The object orientation is not the draw for me, it is > the really good support (albeit only under Windows) and the low cost > of entry. No runtimes, and a minimum of $199 for the standard version > of Visual Studio. Yes, I worked with VS Express for nearly a year and it was no hardship. That version is free. I treated myself to the Enterprise Edition when I realised I could get a whole lot more done (web development, mobiles, the whole nine yards...) $600 for the "Professional" version, which adds a > few nice touches. > > So to me, the COBOL world, in which vendors where accommodating, > welcoming, developer friendly, and interested in their customers has > virtually disappeared, with the possible exception of the new PerCOBOL > people. > > The future looks bleak folks- I sure hate this. But we knew about it 15 years ago... It wasn't like it was a big surprise. People just wouldn't read the writing on the wall. I don't think it's bleak. Not even for COBOL. The chances of making a living as a one trick COBOL pony are pretty bleak, but there is an argument that says we shouldn't have done that anyway... If people are allowed to move off COBOL in a measured way the future can be bright. > > Anyone else getting a different picture of the COBOL world? I would > sure love to hear about it if you are. > > -Paul > > (Passing virtual six-pack over to Pete now... :) Thanks Paul, I needed a beer after that... :-) Pete. -- "I used to write COBOL...now I can do anything."
From: Howard Brazee on 11 Sep 2009 10:00 On Fri, 11 Sep 2009 17:56:51 +1200, "Pete Dashwood" <dashwood(a)removethis.enternet.co.nz> wrote: >> The future looks bleak folks- I sure hate this. > >But we knew about it 15 years ago... It wasn't like it was a big surprise. >People just wouldn't read the writing on the wall. Some people wouldn't read it - but others read it quite accurately. If a genie gives us the ability to predict the future, the story usually has us creating that future, despite everything we do to change it. But I don't see that happening here - we were just powerless to change it. That's not to say that changing it is necessarily smart. There are reasons things have changed that have nothing to do with the stubbornness of mainframers not switching to OO CoBOL. >I don't think it's bleak. Not even for COBOL. The chances of making a living >as a one trick COBOL pony are pretty bleak, but there is an argument that >says we shouldn't have done that anyway... If people are allowed to move off >COBOL in a measured way the future can be bright. CoBOL hasn't been replaced by an other language. But people can no longer expect one technical skill to be sufficient for their IS career. -- "In no part of the constitution is more wisdom to be found, than in the clause which confides the question of war or peace to the legislature, and not to the executive department." - James Madison
From: rtwolfe on 11 Sep 2009 11:03 Paul: The Flexus COBOL Tools are compiler independent and support the latest versions of Windows, and Linux, UNIX and UNIX derivatives as well as OpenVMS, provide client/server and web browser user interface capability as well as a host of modern features. Don't give up on COBOL. Contact us. The best part is that our pricing is very reasonable. send an e-mail or contact us through the web site www.flexus.com. I'll be happy to provide you with answers to your questions. Thanks. On Sep 10, 10:16 pm, Paul <paul-nospamatall.rauler...(a)mac.com> wrote: > I absolutely hate it when Pete's prognostications come true, but COBOL > is becoming near impossible to get and use on mainstream (i.e. Windows > and Linux) platforms. > > A lot of my (personal) software is written in IBM VisualAge COBOL, and > for very good reasons. IBM gave Partners a REALLY big break on the > price, making development feasible, and it doesn't have any runtime > cost. Worked great. But it does not run on any modern version of > Windows, at least not the GUI developer. > > I ported the software to VMS COBOL, which I love. I was thinking this > would give me another 5 to 10 years. But the only machines my customers > can buy that it runs on are Itanium based servers from HP. My customers > are screaming for Windows and Macs refusing to buy Itaniums. > > I have mainframe software, but IBM no longer sells a mainframe for the > "entry level" market. My customers on Flex and MP3K's are running out > of time and have no place to go - but they sure know where they WANT to > go - Windows, Macs, or Linux. > > I have AS/400 software, but the darn platform is going into oblivion - > take it's new name for instance "i". Of course, there is still AIX > COBOL, but that is being pushed hard and heavy by Linux. > > Then you go talk to the existing COBOL vendors today. Microfocus > wants $4K for the NetExpress Developer - and $22K per server. Perhaps > that is *per core* - I don't know, I was so shocked by their price > increase I didn't bother asking. And they have sucked up both AcuCOBOL > and Liant/RM. No options. > > Veryant COBOL sounds interesting, but they want a great deal of money > for their development platform, and their runtimes are "half the cost" > of MicroFocus. $11 per core is not cheap. :/ > > These Alchemy guys that took over NetCOBOL from Fujitsu are arrogant > beyond belief. After just a short interaction with them, I think I > would bite myself in the small of my back before I would send them > money. > > All these vendors seem to have hit upon the idea of making COBOL > "exclusive." Or something. Charge high for it and people will flock to > buy it. Good luck with that. > > The sole exception that *I* have found is PerCOBOL. They seem to be > excited about what they are doing, and lo and behold, they can and will > talk intelligently with a person! Even discuss the benefits of > competing products. I've not yet purchased their development set, but > we may come to mutually happy agreement. I like their compiler, and it > is a *easy* entryway to Java for a COBOL programs. (You guys might want > to check it out if you are doing projects. I'm not affiliated with them > in any way, but just the fact they will talk to you and try to help you > understand their products is a very big plus in my book.) > > But back to Pete's ideas - darn if one of the best options for porting > my COBOL code to something a bit more modern hasn't turned out to be > Microsoft C#. :) The object orientation is not the draw for me, it is > the really good support (albeit only under Windows) and the low cost of > entry. No runtimes, and a minimum of $199 for the standard version of > Visual Studio. $600 for the "Professional" version, which adds a few > nice touches. > > So to me, the COBOL world, in which vendors where accommodating, > welcoming, developer friendly, and interested in their customers has > virtually disappeared, with the possible exception of the new PerCOBOL > people. > > The future looks bleak folks- I sure hate this. > > Anyone else getting a different picture of the COBOL world? I would > sure love to hear about it if you are. > > -Paul > > (Passing virtual six-pack over to Pete now... :)
From: Clark F Morris on 11 Sep 2009 12:02
On Thu, 10 Sep 2009 21:16:15 -0500, Paul <paul-nospamatall.raulerson(a)mac.com> wrote: >I absolutely hate it when Pete's prognostications come true, but COBOL >is becoming near impossible to get and use on mainstream (i.e. Windows >and Linux) platforms. > >A lot of my (personal) software is written in IBM VisualAge COBOL, and >for very good reasons. IBM gave Partners a REALLY big break on the >price, making development feasible, and it doesn't have any runtime >cost. Worked great. But it does not run on any modern version of >Windows, at least not the GUI developer. > >I ported the software to VMS COBOL, which I love. I was thinking this >would give me another 5 to 10 years. But the only machines my customers >can buy that it runs on are Itanium based servers from HP. My customers >are screaming for Windows and Macs refusing to buy Itaniums. > >I have mainframe software, but IBM no longer sells a mainframe for the >"entry level" market. My customers on Flex and MP3K's are running out >of time and have no place to go - but they sure know where they WANT to >go - Windows, Macs, or Linux. > >I have AS/400 software, but the darn platform is going into oblivion - >take it's new name for instance "i". Of course, there is still AIX >COBOL, but that is being pushed hard and heavy by Linux. > >Then you go talk to the existing COBOL vendors today. Microfocus >wants $4K for the NetExpress Developer - and $22K per server. Perhaps >that is *per core* - I don't know, I was so shocked by their price >increase I didn't bother asking. And they have sucked up both AcuCOBOL >and Liant/RM. No options. > >Veryant COBOL sounds interesting, but they want a great deal of money >for their development platform, and their runtimes are "half the cost" >of MicroFocus. $11 per core is not cheap. :/ > >These Alchemy guys that took over NetCOBOL from Fujitsu are arrogant >beyond belief. After just a short interaction with them, I think I >would bite myself in the small of my back before I would send them >money. > >All these vendors seem to have hit upon the idea of making COBOL >"exclusive." Or something. Charge high for it and people will flock to >buy it. Good luck with that. > >The sole exception that *I* have found is PerCOBOL. They seem to be >excited about what they are doing, and lo and behold, they can and will >talk intelligently with a person! Even discuss the benefits of >competing products. I've not yet purchased their development set, but >we may come to mutually happy agreement. I like their compiler, and it >is a *easy* entryway to Java for a COBOL programs. (You guys might want >to check it out if you are doing projects. I'm not affiliated with them >in any way, but just the fact they will talk to you and try to help you >understand their products is a very big plus in my book.) > >But back to Pete's ideas - darn if one of the best options for porting >my COBOL code to something a bit more modern hasn't turned out to be >Microsoft C#. :) The object orientation is not the draw for me, it is >the really good support (albeit only under Windows) and the low cost of >entry. No runtimes, and a minimum of $199 for the standard version of >Visual Studio. $600 for the "Professional" version, which adds a few >nice touches. > >So to me, the COBOL world, in which vendors where accommodating, >welcoming, developer friendly, and interested in their customers has >virtually disappeared, with the possible exception of the new PerCOBOL >people. > >The future looks bleak folks- I sure hate this. > >Anyone else getting a different picture of the COBOL world? I would >sure love to hear about it if you are. To add to your gloom, IBM mainframe COBOL doesn't support decimal floating point, IEEE floating point or the new data types in the 2002 standard. It communicates to Java through a kludge. Somehow the bean counters, not the long run strategists are in charge. The Computer Science idiots and shortsighted bean counters are winning. > >-Paul > >(Passing virtual six-pack over to Pete now... :) |