This just in from Olero

Posts   
1  /  2  /  3
 
    
JimFoye avatar
JimFoye
User
Posts: 656
Joined: 22-Jun-2004
# Posted on: 01-Sep-2004 17:34:29   

The Future of ORM.NET

Dear ORM.NET Customers and Supporters,

Olero Software is no longer able to continue supporting ORM.NET. We apologize for the lack of communication over the last couple of months and any inconveniences this may have caused you.

We still believe in ORM.NET as a valuable tool and, therefore, have decided to make the source code available to everyone via SourceForge.NET. We hope this will allow current customers the ability to continue to use and make improvements to the tool. It will also allow us –time permitting – to make improvements and fix defects.

Please submit bug reports via the SourceForge site. We will still be actively fixing any bugs reported and supporting paid customers We will be working to improve the online documentation. If you are interested in development/CVS access to the ORM.NET source tree, please put in a request! For the latest version and complete source go to: http://sourceforge.net/projects/orm-net/

Sincerely, Olero Software

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 01-Sep-2004 17:59:58   

My sense of OpenSource software is, you get what you pay for.

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39613
Joined: 17-Aug-2003
# Posted on: 01-Sep-2004 18:06:08   

I saw that too, not unexpected though. They abandoned their customers 4 months ago or so, as some people on their forums were really desperate: no-one helped them.

I'm not sure why they abandoned the product. Ok, it's not top quality in some areas but it did have a market, but perhaps not enough sales were rolling in, don't know. They're not alone though. Pragmatier is also out of business and Objectz.net by Mongoose is also discontinued.

I don't think ORM.NET in open source format will be used a lot. The code is clean but not very open to drastic changes, so it will take a lot of effort to change it, effort which can better be spend elsewhere. simple_smile .

What I found weird is that they already uploaded 1.7 to sourceforge in April !.

Frans Bouma | Lead developer LLBLGen Pro
swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 01-Sep-2004 18:22:34   

I hadn't heard that Objectz.net went under. I know that TierDeveloper is still around; I get hammered by emails from them about their new version 4.0 nearly every day. rage

The other day a friend pointed out the risk that you might get hit by a truck and this thing all goes away. Please look both ways before crossing the street. Too many people are depending on you! simple_smile

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39613
Joined: 17-Aug-2003
# Posted on: 01-Sep-2004 18:44:12   

swallace wrote:

I hadn't heard that Objectz.net went under. I know that TierDeveloper is still around; I get hammered by emails from them about their new version 4.0 nearly every day. rage

Ah yes alachi-soft... their resellers are really spam-monkeys indeed. Objectz.net is not officially dead yet, but there isn't much life in it since a few months.

The other day a friend pointed out the risk that you might get hit by a truck and this thing all goes away. Please look both ways before crossing the street. Too many people are depending on you! simple_smile

Yeah, that's the risk indeed... IF we go out of business for whatever reason (not very likely, but you'll never know what kind of misery might happen, financially we're very strong simple_smile ) customers will get all the sourcecode under the GPL. We toyed with the idea of releasing all the sourcecode to customers so the dependency is gone, but it would be a bit too risky for us, releasing our crown jewels for 170euro... at least for now simple_smile

Frans Bouma | Lead developer LLBLGen Pro
swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 01-Sep-2004 19:59:25   

Just went skimming through the ORM.NET source code. Couple of tasty treats there, particularly amongst the utilities (OrmLib). A few good string parsing routines, and a pretty good threaded jobqueue system, though I couldn't find one place in the code where it was used. The SQL handling stuff is interesting but not helpful.

I looked for something regarding making collections visible in designer, or whatever that problem was you were having, but I'm not sure what I'm looking for. You might visit it for clues in that area. Most properties seem to be marked [EditorBrowsable(EditorBrowsableState.Never)] which indicates to me it won't be available in the designer (but what do I know?).

Their templates are stored as embedded resources in an assembly. I guess that would work, but it certainly prevents end-user editing.

There's a whole folder called "do_not_edit". Not sure why. Other than that, nothing embarrasing or unusual, nor much to learn from.

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 01-Sep-2004 20:19:22   

Yep, built a DAL with ORM.NET. The do_not_edit folder is a template that then is populated and placed into your DAL. Makes sense, but slightly offensive. simple_smile

It's hard to believe this was state of the art at one point. The interface is so bereft of features it's now laughable. It's built around returning datasets and datarows, and offers few advanced features.

It would take teams of programmers (working for free?) a year to bring this thing up to the current commercial release of LLBLGenPro.

wayne avatar
wayne
User
Posts: 611
Joined: 07-Apr-2004
# Posted on: 02-Sep-2004 09:33:20   

They're not alone though. Pragmatier is also out of business and Objectz.net by Mongoose is also discontinued.

Yip everybody jumped onto the O R Mapper wagon, but only the best will survive. In the end there will only be a 1 or 2 left and then...micrisoft will come and buy the best one out. Imagine - Microsoft LLBLGen stuck_out_tongue_winking_eye i'm sure Otis will be happy...Financially in any case.

Funny i thought pramatier was a very good O R Mapper...what happened?

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39613
Joined: 17-Aug-2003
# Posted on: 02-Sep-2004 09:46:43   

swallace wrote:

Just went skimming through the ORM.NET source code. Couple of tasty treats there, particularly amongst the utilities (OrmLib). A few good string parsing routines, and a pretty good threaded jobqueue system, though I couldn't find one place in the code where it was used. The SQL handling stuff is interesting but not helpful.

I found it a bit dissapointing really. It was ok, but I did expect a more proper query engine...

I looked for something regarding making collections visible in designer, or whatever that problem was you were having, but I'm not sure what I'm looking for. You might visit it for clues in that area. Most properties seem to be marked [EditorBrowsable(EditorBrowsableState.Never)] which indicates to me it won't be available in the designer (but what do I know?).

simple_smile Well, they use datasets, so I'm not sure if they solved it that way, as datasets can be designed in the IDE without problems. Microsoft is now looking at my example and why it isn't working, for 5 days now, I'll mail them later this week to see if they've fallen off their chair or just ignored the email wink

But I'll have another look. Although it's GPL-ed code, so I won't copy a single bit, it can give a clue perhaps. After all, the one who wrote the best (and only!) article before I wrote mine on ITypedList was the lead developer of Olero simple_smile

swallace wrote:

Yep, built a DAL with ORM.NET. The do_not_edit folder is a template that then is populated and placed into your DAL. Makes sense, but slightly offensive. simple_smile

hehe, I get the feeling you have to surpress the urge to edit that do_not_edit folder wink

It's hard to believe this was state of the art at one point. The interface is so bereft of features it's now laughable. It's built around returning datasets and datarows, and offers few advanced features.

Yeah, and that for 495$ per developer simple_smile . Though their joinpaths are ok, I must admit. For the rest it's pretty basic...

It would take teams of programmers (working for free?) a year to bring this thing up to the current commercial release of LLBLGenPro.

simple_smile indeed. That's also the disadvantage open source solutions have: after a while you get something running, but it takes a lot of effort to make it worth using because all those extra features really make the stuff incredibly complex. The sourcecode for the runtime libs of llblgen pro is 1.4MB of C# already simple_smile . For open source projects, it requires a lot of free time for a long time to stay on top or even get there. Everybody in the open source O/R mapper world for .NET wants to make the next Hibernate, but forget that the guys who wrote Hibernate worked for a loooooong time on the platform and do that full time now. You can't get there with just a few hours a day for 4 days a week.

Frans Bouma | Lead developer LLBLGen Pro
Otis avatar
Otis
LLBLGen Pro Team
Posts: 39613
Joined: 17-Aug-2003
# Posted on: 02-Sep-2004 09:55:41   

wayne wrote:

They're not alone though. Pragmatier is also out of business and Objectz.net by Mongoose is also discontinued.

Yip everybody jumped onto the O R Mapper wagon, but only the best will survive. In the end there will only be a 1 or 2 left and then...micrisoft will come and buy the best one out. Imagine - Microsoft LLBLGen stuck_out_tongue_winking_eye i'm sure Otis will be happy...Financially in any case.

stuck_out_tongue_winking_eye simple_smile It will be interesting what will happen in the future in this market. The niche will get bigger, and of the 5 or so top tools 2 will be the big players and the rest will live in the margin. I'm not sure if Microsoft will buy one soon though. I don't have the feeling Microsoft is even convinced O/R mapping is useful.

Funny i thought pramatier was a very good O R Mapper...what happened?

They leaked money from the start and didn't innovate. Mats explained it on his weblog: http://www.matshelander.com/Weblog/DisplayLogEntry.aspx?LogEntryID=11

Frans Bouma | Lead developer LLBLGen Pro
wayne avatar
wayne
User
Posts: 611
Joined: 07-Apr-2004
# Posted on: 02-Sep-2004 10:44:47   

What are the main competitors to LLBLGen?

It seems like a very difficult market...especially because alot of people don't yet believe in O R Mapping and there are so many small whannabe competitors.

Personaly i don't see O/R Mapping as such a new technology, the principle of having an object get your data for you and having properties that are mapped to fields are not new at all, just look at TTable or TQuery component in Delphi 1-6 ,even a datatable in .Net is doing some of that just in a much more primitive way. The only difference with an O R Mapper is that it generate it for you (isn't that coolwink ) and creates a derived object that is unique to a table in your DB.

What is the fuss about?

I was already doing simular types of logic 3-4 years ago...but coding it by hand. - How else would you do OO Tiered applications?

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 02-Sep-2004 16:30:56   

How good is this market? It's good. Let me tell you. (This may be a continuation of the thread listed in a previous post here. Stick with me while I set the stage...)

I'm in Oklahoma City, and I'll be the first to tell you that people here are the last to jump on a bandwagon. When a fast-moving freight train full of gold comes by, they're the ones running behind it still trying to decide whether all that gold makes it worth jumping on board. (Can we eat it? Will it hurt us? I hear it's awfully heavy... I want to scream, it's GOLD you dumb*ss!)

I do contract work, and I'm coming on the end of my current effort. I've been looking around, and all I see are VB6 jobs. Here we are, three (four?) years after the release of VS.NET, with the third edition looming, and everyone I know is still working traditional ASP and VB6.

In each of my last two contracts, where I finally found VS.NET places, I introduced them to the idea of O/M tools (LLBLGen specifically). They didn't know such things existed.

(Still setting things up here...) For the last three years labor has been cheap, cheap, cheap, and I doubt it's just in this market. You know, the Romans knew of the power of steam, and even had several (lousy) steam-powered novelty objects. But they never viewed steam as a labor-saving method because labor was cheap, cheap, cheap.

So, where are we going here? Why is Frans about to be a gazillionaire?

  1. The chatter now is, "Say, that VS.NET 2005 might make it worth beginning to transist all this ASP and VB6 to .NET code." The masses are just coming to .NET, believe it or not, and they will continue to open the marketplace for add-ins.
  2. The labor market is turning. Labor two years from now will not be as cheap, and tools that save contractor hours (product positioning point, Frans) will be sought out.
  3. The O/M market has shaken out, and the strong have survived. There was a point some 18 months ago when everyone thought they could write such a tool, and everyone tried. Everyone thinks so because at one point everyone has had to write a DAL for an n-Tier project. The products that still exist, TierDeveloper, Deklarit, and LLBLGen, have a maturity that will act as a barrier to the amateur (I can do that!) developer for the next two years. Frans is perfectly positioned with a mature, full featured product.
  4. The concept of n-Tier is starting to get noticed because of Microsoft's databinding tools in VS.NET. People who have been doing .NET for a few years were able to realize a significant labor savings just by going to .NET, but are just starting to find out they can save even more with packaged-knowledge tools like O/M products.
  5. Microsoft dropped the ball with ObjectSpaces and the delay of WinFS. They whetted people's appetite for better data-handling tools, and then told them "someday..." That someday is now, if they find out about these third-party tools.

If you read the above and find me contemptuous of most development shops, it's because I am. Let's face it. If you write in .NET, congratulations, you are in the minority amongst developers. If you know of O/M tools, congratulations, you work in the top tier of development shops.

Customers don't demand quality code. They demand a working product at a low price. If I can build it with four $20/hour people using scut tools I'm not interested in a $75/hour guy even if his tool costs only $250. The challenge is to get LLBLGen in the hands of the $20/hour guy, to get him up to speed on these tools. That market is ten times as big. I compete, as a contractor, not against the quality shops, but against the many, many, many "let's build it in Access" shops. Those guys are dying away, not because the customer demands it, but because Microsoft has been relentless in the marketing of .NET and getting it in the hands of developers (like with the upcoming Preview editions). You must do the same.

The O/M market is about to explode. Raising labor costs, more developers in the .NET arena, a mature market, and no Microsoft competition on the horizon. Looks good. It has a natural cap because of the investment of knowledge required to use the tools, but there are ways around that.

The challenges for LLBLGen:

  1. Continue to provide the consistently high level of customer support that exists now. Let's face it, you're only one man. Start looking into methods for outsourcing support.
  2. Get the tools in the hands of the twenty buck guys. It's not about the cost of the tool, it's about the investment of time required to overcome the conceptual hurdle of using it. Why use it, how do I use it, how do I convince my boss/customer I should use it, which one should I use, etc. Most twenty buck guys say, "if it didn't get bundled with VS.NET it must be crap and I don't need to learn it." This is a tough hurdle. Microsoft sometimes allows companies to ship samples of their tools with VS.NET CDs. You should seriously look into it.
  3. It's a one-trick pony. A hell of a pony, but it's just one. I wonder what related-product ideas are on the horizon. InterfaceGenPro? SQLGenPro?

The truth about LLBLGen and Microsoft: The only reason Microsoft would buy out Solutions Design is to hire Frans, to keep him from joining a competitor or to have him build something at the OS level. He is not their competitor, nor will they build LLBLGen into VS.NET. It's not how they operate. For an example, look at the recent acquisition of LookOut. They wanted to hire the guy to have him build search tools in Longhorn, and LookOut was his resume. They re-released the tool, once they'd bought the company, under the Microsoft name, but even the original developer says that it won't be continued or supported. They wanted the man, and they got him. When Microsoft sees that Frans could help them with ObjectSpaces or WinFS, they'll make an offer. A good offer. And if he takes it then I'll stop using LLBLGen.

I feel better. Nothing like a few cups of tea, a keyboard, and a small audience to lighten your soul.

simple_smile

wayne avatar
wayne
User
Posts: 611
Joined: 07-Apr-2004
# Posted on: 02-Sep-2004 17:21:34   

hmmm...nice piece of writing there Bob wink , are you sure that you are a programmer?stuck_out_tongue_winking_eye

Your spelling and writing abilities are very good.

Most programmers i know can't spell properly and type their letter (not refering to chars) as they type in the dev tool Using Pascal Case And Using American Spelling Like Color instead of Colour.

So MS wants the man not the tool...

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 02-Sep-2004 18:14:00   

hmmm...nice piece of writing there Bob , are you sure that you are a programmer?

Thanks. simple_smile As to being a programmer, I suspect there are plenty of people who would contend that I am not, especially compared to others! Perhaps that's why I'm always looking for the best tools and components; I want to look good from the work of others. And why not stand on the shoulders of giants, and for only $249?

Funny, I was just talking career options with a friend this morning. She was saying she'd like to go back to teaching, and I said I would like to go back to writing. I did that for a while, co-hosted a local weekly radio show about computers for a few years, co-authored a computer book, other things, but here I am writing software again. Go figure.

Frans will reach a point, likely soon after he finishes the VS.NET 2005 version, when he has to make decisions about growth, product depth, and broadening the customer base. I contend that the twenty buck guys are the key, but they come at a high price. Giveaways, product promotion, vastly increased product support, and more.

The trick is whether he's out to produce the best O/M product available, or out to insure that his vision of increasing programmer productivity through intelligent tools is accepted and understood by all developers. What do you sell, a product or a vision? If it's the former, it's done. But the later opens new vistas, presents new product ideas, and creates new demands, the first of which is a clear articulation of the vision. Microsoft set out to put a computer on every desk and in every home, not to make money selling Word and Excel. They succeeded at both because of the vision, not because of the quality(?) of the products.

The demands (and costs of obtaining the love) of the twenty buck guys is offset by the revenues generated by additional products that feed the vision.

How cool is that?

And what does that have to do with my own career path?

I need another up of tea...

wayne avatar
wayne
User
Posts: 611
Joined: 07-Apr-2004
# Posted on: 02-Sep-2004 20:04:25   

when he has to make decisions about growth, product depth, and broadening the customer base.

Volume of sales are always important.

I contend that the twenty buck guys are the key

hmm, it seems like i am falling into that category - That is if i convert my hourly rate to US$.wink

Believe me $ 20 per hour here is MEGGA BUGS!!

jeffreygg
User
Posts: 805
Joined: 26-Oct-2003
# Posted on: 02-Sep-2004 20:50:39   

Enjoyed your "article".simple_smile I agreed with most of it, but a couple of thoughts:

swallace wrote:

  1. Continue to provide the consistently high level of customer support that exists now. Let's face it, you're only one man. Start looking into methods for outsourcing support.

Definitely don't agree here. I have not once had a pleasant experience with outsourced support, have you? Assuming Frans' desire is to maintain high-quality support (his trademark), he can't outsource. The best support lies in the hands of the creator, or those close to him, and that means carefully chosen employees. Outsourcing reduces support to the call volume and response times. These metrics are anathema to good support; They put the focus of the tech on meeting quotas instead of solving problems. Of course, this whole conversation makes a lot of assumptions about the level of growth Frans wants. simple_smile

  1. Get the tools in the hands of the twenty buck guys. It's not about the cost of the tool, it's about the investment of time required to overcome the conceptual hurdle of using it. Why use it, how do I use it, how do I convince my boss/customer I should use it, which one should I use, etc. Most twenty buck guys say, "if it didn't get bundled with VS.NET it must be crap and I don't need to learn it." This is a tough hurdle. Microsoft sometimes allows companies to ship samples of their tools with VS.NET CDs. You should seriously look into it.

I agree mostly. ORM needs to be in the hands of everybody not just developers "in the know." Not just because it's a time saver per se, which it is, but because it's the next step. It's a level of abstraction that must be there - and will be there, IMO. But, I think it's education, not marketing. Part of the problem is that those twenty dollar an hour guys don't created well abstracted programs with tiers and layers and so forth. They do direct binding to the data source and are very pleased with themselves. ORM is not even relevant to these guys. Yes, we need to abstract the data source from these guys, but they aren't going to be the ones to do it. It's going to take someone like MS or someone else to introduce this concept and, yes, force it on them.

ORM is inherently a tool that will stay in the province of the thoughtful developer, instead of the do-it-quick programmer until it is adopted by a major player. This is true especially due to the current method of implementing ORM: code generation. Until the objects are available on a dynamically generated basis, I don't see the do-it-quick guys using it. It just doesn't fit into their development environment and methodology. Setting up a whole other tool and directories and projects and so forth is not something they will think is worth while - especially because of the strong data-binding support in VS.NET. When the objectification of the data source is automated, then we'll see it come into the main stream. Until then, it is only worth while to the organized, thoughtful developer who has the opportunity to create well-architected solutions.

  1. It's a one-trick pony. A hell of a pony, but it's just one. I wonder what related-product ideas are on the horizon. InterfaceGenPro? SQLGenPro?

This may be just me, but I tend to stay away from products that "do it all". Too much vendor lock in, not enough flexibility in design and implementation, and an orders of magnitude level of scary in terms of liability. Not to mention the subconscious feeling of how can one thing do everything well? I know this just may be me, but...

Jeff...

Aglaia avatar
Aglaia
LLBLGen Pro Team
Posts: 535
Joined: 07-Sep-2003
# Posted on: 02-Sep-2004 21:02:19   

swallace wrote:

How good is this market? It's good. Let me tell you. (This may be a continuation of the thread listed in a previous post here. Stick with me while I set the stage...)

Snip setting the stage and why Frans is about to be a gazillionaire

The challenges for LLBLGen:

  1. Continue to provide the consistently high level of customer support that exists now. Let's face it, you're only one man. Start looking into methods for outsourcing support.

Well, actually he isn't completely alone in this company simple_smile The forum just is a bit too technical for me, so I'm the Unknown One doing all the questions from potential customers. I'm also the one handling the little marketing we do, and the one making sure Frans doesn't spend too much time doing customer support. The minute he is, we'll have to hire someone.

  1. Get the tools in the hands of the twenty buck guys. It's not about the cost of the tool, it's about the investment of time required to overcome the conceptual hurdle of using it. Why use it, how do I use it, how do I convince my boss/customer I should use it, which one should I use, etc. Most twenty buck guys say, "if it didn't get bundled with VS.NET it must be crap and I don't need to learn it." This is a tough hurdle. Microsoft sometimes allows companies to ship samples of their tools with VS.NET CDs. You should seriously look into it.

The 'why, how, which one' stuff I'm going to think about, thanks. We did try to emphasize the time-saving aspect in the texts, but the rest may need tweaking. As for the shipping of samples - we're going to try, but according to Frans it's hard to get Microsoft to do such a thing. Furthermore, we did banner a lot in the beginning, but that didn't seem to generate more sales at all. And even being a marketeer I'd say you shouldn't invest too much in advertising if it doesn't earn much more profit. Nonetheless, we do plan on bannering again, even if it's just for your twenty buck guys to see there could be a product helping them. I've got the feeling it takes time to allow programmers getting used to the idea that 'their own' is always better.

  1. It's a one-trick pony. A hell of a pony, but it's just one. I wonder what related-product ideas are on the horizon. InterfaceGenPro? SQLGenPro?

We're thinking about the next CoolTool, but at the moment Frans is working on perfecting LLBLGen ...

The truth about LLBLGen and Microsoft: The only reason Microsoft would buy out Solutions Design is to hire Frans, to keep him from joining a competitor or to have him build something at the OS level. He is not their competitor, nor will they build LLBLGen into VS.NET. It's not how they operate. For an example, look at the recent acquisition of LookOut. They wanted to hire the guy to have him build search tools in Longhorn, and LookOut was his resume. They re-released the tool, once they'd bought the company, under the Microsoft name, but even the original developer says that it won't be continued or supported. They wanted the man, and they got him. When Microsoft sees that Frans could help them with ObjectSpaces or WinFS, they'll make an offer. A good offer. And if he takes it then I'll stop using LLBLGen.

simple_smile If he takes it, there will be a lot of discussion around here... He shouldn't work for any boss, it takes away his freedom of brilliance.

Thanks for your input. I'm going to think about getting the twenty buck guys interested simple_smile

Aglaia

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39613
Joined: 17-Aug-2003
# Posted on: 02-Sep-2004 21:09:17   

wayne wrote:

when he has to make decisions about growth, product depth, and broadening the customer base.

Volume of sales are always important.

True simple_smile

I contend that the twenty buck guys are the key

hmm, it seems like i am falling into that category - That is if i convert my hourly rate to US$.wink Believe me $ 20 per hour here is MEGGA BUGS!!

whoa simple_smile Here 17EUR / hour is what you get as a normal employee...

btw, I pressume you wanted to say 'Mega Bucks!!'? Or "with that amount of money, you pay peanuts so you get monkeys and thus bugs" ? wink

Frans Bouma | Lead developer LLBLGen Pro
swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 02-Sep-2004 21:15:59   

Aglaia, good to meet you. I had no idea Frans had a partner, and that you were lurking. I notice that was your first post. I feel proud to have drawn you out. wink

We're thinking about the next CoolTool, but at the moment Frans is working on perfecting LLBLGen

More power to him! We love it!

The minute he is, we'll have to hire someone.

I'm glad to hear that. There's this Catch-22 that the high level of support draws customers, but that the product suffers because of the high level of support, sending customers away.

That, to me, is the key to the twenty buck guys. They are fools (and wayne, you are not one of them) who must have their hands held all the way. You can lead them in through advertising, but you can't draw the sale out of them because they simply haven't overcome the conceptual hurdle of how the product benefits them and how it's intended to work.

It seems that the answer involves a) extended customer support, b) increased documentation, white papers, tutorials, and 3) structural changes to the product to make it easier for the twenty buck guys. (I have no idea what those would be. The simple answer is wizards, the more complex answer is likely found in Microsoft's Task Pane and SmartTags. Either way, a study of how twenty buck people use (or confuse) the product is in order).

but according to Frans it's hard to get Microsoft to do such a thing.

I'm sorry to hear that. I suspected it would be very expensive as well. Myself, I wouldn't spend a dime more on advertising until the product is a factor of ten more intuitive. That's not a slam, just a recognition, as you say, that not everyone understands the need. That's your real challenge. Sharing your vision, making it simple to get.

Thanks for listening!

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39613
Joined: 17-Aug-2003
# Posted on: 02-Sep-2004 21:20:47   

jeffreygg wrote:

swallace wrote:

  1. Continue to provide the consistently high level of customer support that exists now. Let's face it, you're only one man. Start looking into methods for outsourcing support.

Definitely don't agree here. I have not once had a pleasant experience with outsourced support, have you? Assuming Frans' desire is to maintain high-quality support (his trademark), he can't outsource. The best support lies in the hands of the creator, or those close to him, and that means carefully chosen employees. Outsourcing reduces support to the call volume and response times. These metrics are anathema to good support; They put the focus of the tech on meeting quotas instead of solving problems. Of course, this whole conversation makes a lot of assumptions about the level of growth Frans wants. simple_smile

There is always a limit on what you can handle of course. What's funny is that sales are going steady upwards but the amount of support requests is stable for at least 6 months now. simple_smile So handling it is very doable.

Growth is something that has to be logical: just increasing the amount of employees is not necessarily what I see as a plus or a thing to achieve. What I learned in the last couple of years is that development isn't going that much faster when development is outsourced to other people, because keeping everyone in sync with what has to be done takes time.

However it will be inevitable one day I think, especially if the tool grows very big with a lot of different functionality, often a stage that is reached after a few years (I'm not sure who said it but it came down to: "every application eventually allows you to read email" simple_smile ).

  1. Get the tools in the hands of the twenty buck guys. It's not about the cost of the tool, it's about the investment of time required to overcome the conceptual hurdle of using it. Why use it, how do I use it, how do I convince my boss/customer I should use it, which one should I use, etc. Most twenty buck guys say, "if it didn't get bundled with VS.NET it must be crap and I don't need to learn it." This is a tough hurdle. Microsoft sometimes allows companies to ship samples of their tools with VS.NET CDs. You should seriously look into it.

I agree mostly. ORM needs to be in the hands of everybody not just developers "in the know." Not just because it's a time saver per se, which it is, but because it's the next step. It's a level of abstraction that must be there - and will be there, IMO. But, I think it's education, not marketing. Part of the problem is that those twenty dollar an hour guys don't created well abstracted programs with tiers and layers and so forth. They do direct binding to the data source and are very pleased with themselves. ORM is not even relevant to these guys. Yes, we need to abstract the data source from these guys, but they aren't going to be the ones to do it. It's going to take someone like MS or someone else to introduce this concept and, yes, force it on them.

You're absolutely right: education of what O/R mapping is, what it can do, is key, and the vast majority doesn't know about this technology. I also don't see it as 'O/R mapping is the solution', because that's not correct I think. I think a tool which uses O/R mapping (and other techniques) can be an answer to the data-access-problem a lot of developers are facing. When a lot more people will understand that there are more answers to that problem then old datasets is hard to say... simple_smile

Frans Bouma | Lead developer LLBLGen Pro
wayne avatar
wayne
User
Posts: 611
Joined: 07-Apr-2004
# Posted on: 02-Sep-2004 21:22:59   

whoa simple_smile Here 17EUR / hour is what you get as a normal employee...

btw, I pressume you wanted to say 'Mega Bucks!!'? Or "with that amount of money, you pay peanuts so you get monkeys and thus bugs" ? wink

ha ha! smile funny....fransie.

Junior developers get +- R 5000 ($ 724.63) a month here. Normal developers get R 8000 + ($ 1159.42) a month here. Senior developers get R 14000 + ($ 2028.9sunglasses a month here.

I guess the developers in this country gets viewed as cheap labour by International standards.

jeffreygg
User
Posts: 805
Joined: 26-Oct-2003
# Posted on: 02-Sep-2004 21:28:36   

Mrs. Frans? Great to meet you! Welcome!

At any rate I just don't think the lower tier (I don't say that in a deprecating way) of programmers are the target market. At least not until, as swallace said, the tools is orders of magnitude easier - which I think begins with automating the generation of the code. If that's even possible. I think the two big things that will be required to get the twenty buck guys on board are:

  1. Auto-generation of the code
  2. Design time support for databinding.
  3. To some extent, the adoption by a major player

<Edit> It's gotta compete with the current ease of development in the VS IDE for the lower tier programmer. Otherwise, there's no draw. Jeff...

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 02-Sep-2004 21:30:20   

jeffreygg:

Enjoyed your "article".

I do froth on, don't I!

I have not once had a pleasant experience with outsourced support, have you?

You're 100% right here. I love the support here, but I also want the product to grow and thrive. If we can't clone him, the product must be made easier to use, more intuitive, a more natural fit for the way a "simple" programmer works (God help us all.) That way they aren't begging for simple answers to simple questions, the company makes more money (is viable), and the tough questions get answered (because they're the only ones that get asked).

and that means carefully chosen employees.

Yep, I should not have used the dirty word "outsource". I should have said, "other persons whom you trust to provide the same high level of support."

It's going to take someone like MS or someone else to introduce this concept and, yes, force it on them.

You're likely right here. That's certainly been the case in the past. However, I think this product is capable of becoming a vanguard, a breakthrough product that brings intelligent time-saving tools to the average user. I think we're both saying it's possible to create a tool that even boneheaded developers appreciate, but your above sort of implies that only Microsoft can build easy to use tools. I don't agree, and I think the potential pay-off is enormous.

Until then, it is only worth while to the organized, thoughtful developer who has the opportunity to create well-architected solutions.

Gee, I never knew you were such a bigoted pig! simple_smile Just kidding! simple_smile

but I tend to stay away from products that "do it all".

Yep, me too. Coding committees create cr*p. The worst are the tools vendors who buy the small shops, cobble together a lousy product and call it a "suite." ComponentOne comes to mind.

Still, if the vision relates to intelligent time-saving tools, more than one is appropriate...

But, who the heck am I to say?

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 02-Sep-2004 21:37:38   

[quotenick="wayne"]

whoa simple_smile Here 17EUR / hour is what you get as a normal employee...

These things are so relative. I'm sure Jeff in California makes way more than me in Oklahoma, but I suspect my mortage and food bills are significantly cheaper.

A friend and I were lamenting our mutual fate recently, comparing it to experimental chickens who peck at a button and food comes out. We peck at our keyboards, and ...

swallace
User
Posts: 648
Joined: 18-Aug-2003
# Posted on: 02-Sep-2004 21:42:45   

I'm spent on this topic.

Anyone know any good jokes?

1  /  2  /  3