Posted 05 July 2001 - 01:31 PM
Posted 06 July 2001 - 07:15 AM
I know this is re-hashing old discussions (public domain bootstraps!!) but there are obviously people out there who *like* to be restricted in this way and I'm slightly curious as to why.
Posted 06 July 2001 - 12:02 PM
We have a situation here where PROIV COULD provide all the tools we need in one package. We need to encourage them to do so, and to do it RIGHT.
1. We need to tell them what tools we require.
2. The tools HAVE TO WORK.
3. They cannot have all the 'unease' of use that plagues Developers Studio.
4. The tool set has to give us the ability to do EVERYTHING we need, including access to the deep dark 'secrets'.
Yes I want all this in ONE interface, from ONE vendor. Then I can jump up and down their desks in righteous anger when it doesn't work.
Only a fool would take the path of totally restricting access to their product. History has shown that open systems succeed, closed propritary systems fail. EVERY TIME. WITHOUT FAIL. Hands up anyone who thinks the PROIV folks are fools???
I voted for the complete system from one vendor. If/When it doesn't meet all the above criteria, then and only then do I want to have to turn to ancilliary vendors for the tools I need.
Yes, I do want a perfect world.
Posted 06 July 2001 - 02:40 PM
A single development environment that PROIV can concentrate it's development and support resources on
Spread support for all the flavours of development and restrict the ability for PROIV to provide new functionality in the general product and therefore restrict business opportunities for PROIV customers
Lets try to think outside the box a little bit and make sure that our strategy is right. I for one do not want PROIV to stand still
Posted 06 July 2001 - 05:33 PM
The poll is quite adequately stated.
Obviously Neil doesn't think anybody outside Pro-IV capable of making a decent tool. But Pro-Aide was not developed by Pro-IV, although they now own it. No one asks Pro-IV to 'spread its resources' by developing a number of tools. Just one good one would suffice.
The question here is quite simply: does anyone think that the tools supplied will cover all his/her needs, as Pro-IV seems convinced it will. We will have to see that. But as there are many ways to use an environment I would be hugely surprised if anyone can come up with a final solution.
Probably Pro-IV would be better of by devoting its resources to creating an unsurpassed stable, consistent, extendable environment. We all will be quite pleased with a reliable, bugfree kernel. I for one am quite happy to write my own environment (again).
So indeed: public bootstraps, for VIP up front.
Posted 07 July 2001 - 03:00 PM
Yes I would like to see ProIV get it right first time and from what I hear version 5 seems to be promising. I would, however, be _very_ surprised if the one and only development environment provided me with all the tools I will ever need to work on every site, including (to pick just one example) tailored configuration management utilities which will suit a customer site of any size!!
I don't think I would expect any company to be able to anticipate _all_ my needs, present and future, with 100% accuracy, and I doubt ProIV has acquired a crystal ball in the last couple of months. What really is the problem with having more than one development tool? Why shouldn't users be allowed to write their own utilities? These aren't unreasonable questions, just ones that always seem to get skirted and never answered, except with more questions.
Glenn's answer just seems to me to be rather confusing - how many time did you change your mind about the question while writing that? You start off saying it should all be contained in one product (something I have already said I don't think is 100% possible with any product), then you say 'Only a fool would take the path of totally restricting access to their product', yet this is exactly what ProIV seem to be doing by the restrictions that are promised (threatened?) with version 5 and VIP.
Now don't get me wrong, this is not meant to be a criticism of version 5 or VIP in any way - all indications are that version 5 is on the right track and I'll reserve judgement of VIP until I've actually used it, but it does seem that nobody is very keen to answer my original question...
Posted 08 July 2001 - 07:13 AM
You will be able to develop your own tools through the
API that we will provide with VIP.
This approach will mean that once your tools have be converted to use the API they will require no further changes if PROIV changes the shape or storage of the source code.
I am still waiting for the developer community to post examples of tools outside search and replace that are required regularly. If there is a general consensus for a particular feature we will undertake to include it within VIP.
The API is being developed by the HRS developers not by the
VIP team. This is because the HRS team use a lot of source code manipulation tools as well as pre production processes.
There needs to be a bit of give and take here. The development community should get involved in making sure that VIP is going to be right. Version 5.0 of PROIV
will support all the existing development environments but
new functionality (requiring boot extensions) will only be supported by Developer Studio and VIP (when it is released).
We will not extend PRO-AIDE, Native or SL above making them
compatible with V5. Working together and making suitable changes to the design will help us all.
Posted 08 July 2001 - 09:08 PM
OK, here's an example of functionality that goes beyond just search and replace and is vital to my current client.
I've spent a long time developing a set of functionality to support conversion of my client's customers' databases from ISAM databases (across multiple platforms, including PRO-ISAM, RMS on VMS, and C-ISAM) to Oracle and/or SQL-Server. It's designed to build PRO-IV functions on the fly to dump the contents of their databases from any one of several versions of the database and then re-load it into Oracle, potentially on a different platform, under a different operating system, on any one of several versions of the product, along with re-building the RDBMS's attendant procedures, packages, triggers etc. It works well, can merge or de-merge files, and is almost entirely generic, meaning that it can be used with relative ease to convert any customer's database to the new 'improved' Oracle version. BUT if you suddenly change the entire method of creating PRO-IV functions then my client has a whole swag of re-writing to do on a complex set of functionality, which will generate large amounts of cost for little benefit (it works now, why fix/change it?)
I also personally use my own development system on client sites, as I find it much easier to use than Studio - it automates a lot of the processes that Studio doesn't, and also provides version control. Also, because I and my clients have access to the source, we can easily make changes to cope with new requirements and generating standard code.
I do appreciate the idea of an API, but in a sense that's what the bootstraps provide already - a method of getting to the bottom layer of PRO-IV that works well. Studio enforces some standards (I won't say 'contains some bugs') that stop the kernel from behaving in the way it was written to behave - try setting an item of text (say 'Text') in a format to Right aligned, length 20, height 1 in Studio and watch the length and height disappear before your very eyes! This is aggravating in the extreme because it means you can't use Studio to do what you really want to do.
I know that we haven't seen VIP in all its glory yet, and indeed I'm slavering at the mouth with what I've heard about its potential improvements. But to suggest that the rest of the PRO-IV community is incapable of producing top-quality, stable, and productive tools to assist with development is ridiculous - neither you nor I, nor anyone else out there for that matter, has a monopoly on innovation or talent.
So - unless the API is foolproof and TRANSPARENT (i.e. it lets me modify ANYTHING in the bootstraps), which I reserve my judgement on until I've seen it, please publish the bootstraps with 5.0. Issue warnings, tell us the traps, but don't stop us (the developers) from producing good, useful tools to make our jobs easier and (hopefully) our employers able to make good profits.
Posted 09 July 2001 - 12:15 AM
I develop programs for just my company. I own a medium, sized wholesale bakery.
I'm 100% for 5.0 if the fact that I've a service contract lets me get it for having a service contract. Face it a service contract is only good for being able to get with-in version upgrades.
I will NOT touch 5.0 unless it is an upgrade covered by my service contract.
So this is a win-win deal for me. I'll continue to buy service contracts to support pro4 development and use 5.0
or I won't purchase contracts in the future if 5.0 is a pay for upgrade. 4.6 will be my final pro4 version.
The last 2 times I was in Las Vegas I did not bet any $$.. Not even a .25 slot machine.
I will bet that 5.0 will be a pay for upgrade. Who wants to bet a cup of coffee or tea or a beer?
Posted 09 July 2001 - 08:09 AM
I wonder why is it such a big issue to restrict bootstrap access. I would hazard a guess that the bootstrap files are now impossibly difficult to navigate, due to the extra functionality of VIP. I would imagine that the number of Cross Reference files to provide the different views of functions must be quite substanstial.
So is it true to say that the reason ProIV is going to restrict access to the bootstrap files, is because VIP is quite a complicated development tool?
Now I don't mean that as a criticism. I just think that main bootstraps should be as simple as possible. And that everything should be editable (is that a word?) from them. Is there any technical reason why this is not possible?
Then RAD tools such as VIP should be extra and sit on top. Other companies could then get in on the act and write some RAD tools too. ProIV could even support these new RAD tools by publishing in advance changes of any of the bootstrap files.
Why does ProIV feel so responsible for other development environments? If bootstrap files are published then it would be up to the individual companies to update their tools.
Why does ProIV feel it would be such a big step forward to restrict us all to one development environment?
Why does ProIV feel that they would be restricting themselves by having more than one development tool?
Posted 09 July 2001 - 08:46 AM
DS is a vast improvement over the native, ProAide and other development environments in that it does some code validation prior to function generation. It also prevents some of the silly mistakes that inexperienced coders make during function development. Unfortunately, in my own opinion, not enough thought was put into the user interface. Some commonly used options are buried away so that the user has to drill down several levels to get to them.
From what I have seen of VIP so far it would appear that a lot more thought has been expended in making the interface usable. All of the commonly used items are just a single click away. The logic editor has even more validation than DS, so there should be fewer regen/coding correction cycles.
Posted 09 July 2001 - 08:53 AM
From everything I've seen and heard, version 5 and VIP will progress the language in the right direction, which is what we all want.
However, it is not possible for one company to produce a development environment suitable for all their customers needs. ProAide came into being because ProIV had failed to deliver a suitable editor. It was one of ther first editors to display (Updates, cycles or whatever they are called now) in a paging screen, a concept which is seen in Developer Studio and VIP. Would ProIV have come up with this idea (this is a genuine question, not a sarcastic comment) ??
I've heard a rumour that you are going to allow access to the new bootstraps using API (can you confirm this). That's all very well but how many people are comfortable with this solution ??
Utilities are usually written by programmers to save time and get information, therefore they need to be developed and run quickly and efficiently - will this be possible using API ?
Posted 09 July 2001 - 12:36 PM
If you want to write a development environment fine! but it will lag the technology and may never achieve unilateral usage. With one environment we can concentrate on providing common training, quality documentation and developer forums that have the same terms of reference.
The API will enable you to convert any of your existing tools to work with the VIP boots. All relevant data will be exposed for all source code objects with automated updates triggered by the API to maintain the Xrefs and relationships within VIP. The API model will be published and will not required non-disclosure etc.. and will guarantee a future proof solution to user developed tools.
We are about to start the design for the API and these discussions are of great help in making sure that we provide the sort of facilities that the developer community require.
Posted 09 July 2001 - 01:05 PM
2. I do think you should concentrate more on getting the kernel perfect, before worrying about VIP.
3. Is ProIV actually listening to us the developers? Take a look at the poll results so far!
Posted 09 July 2001 - 01:52 PM
To clear things up for me, please answer these. (You are only allowed to answer Yes or No because I'm not big on freedom of speech).
Will I be capable of writing my own development environment eg. Pro-Aide?
Will I be able to use my environment to develop Pro-IV and not rely on VIP?
Will the Lions beat Australia in the final test?
I will be quite happy to maintain it, train users in it, accept responsibility for it and of course, market it.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users