Sponsorship needed for new OS/2 web browser

[Roderick Klein]

Good evening,

thanks for the words, Roderick, but I think this brings me no step further. But that does not matter.

Yes - I am of the opinion, that ecomstation failed because of mismanagement, sadly. And yes, the Situation about the "Browser of the future" reminds me of the Situation "the ACPI of the future of eCS 2.0".

For everbody reading this let just expand a bit more on this discussion about eComStation mis-management.
Because I think its time to set some misconceptions straight.

Sigurd wrote:
"Yes - I am of the opinion, that ecomstation failed because of mismanagement, sadly. And yes, the Situation about the "Browser of the future" reminds me of
the Situation "the ACPI of the future of eCS 2.0".

Without the whole eComStation project there would have not been OS/2 and it would have been long dead. I can understand from a customer perspective the product when it came to ACPI did not deliver what you where expecting off it, sorry about that. I do not think its rocket science to figure out however that both eComStation and ArcaOS mostly the same people where working at Arca Noae as the people that *helped* me at Mensys at the time. The guys at Bitwise works, Steve Levine, David Azarewicz ans Alex Taylor.

I guess its also no surprise that the research and development budget at Mensys never was not millions of Dollars for OS/2. That is/was no secret at all. I remeber the discussions in the public forums about how to manage a project. But with eCS we never had the budget for a large development team or a lot of project managers. Welcome to reality of keeping OS/2 moving forward with all compenents attached to it from browser down to all the drivers. 

When it came to ACPI and it not working I think its fair to say Pasha from the Ukrain did his very best. But ACPI is most likely one of the most complex drivers ever written for OS/2 outside of IBM. Anybody can say about eCo software and Eugene what you want, but eCo software got more projects off the ground such as ACPI and Panorama. Was it of a good quality ? Perhaps not what everybody wanted/was expecting but we would most likely not even be having this discussion in 2017 with the efforts ..

Later things did improve on ACPI  when David Azaricz stepped in at Mensys. And was BTW the person who invited David to start helping us out at Mensys at the time. But finding a developer that can write and debug that sort of code low level code is certainly not easy to find.

Mensys hired a company from the Ukraine to work on the Intel and Realtek OS/2 driver ported from Linux (this was not eCo software). They made the first Intel Gigabit driver based on Linux kernel sources at the time, that worked pretty good. When they started working on the Realtek OS/2 driver things started out ok, but ended in disaster of nothing short. At the company they had three project managers failed on trying to fix this trapping Realtek driver and 2 developers spent close to 6 months on this. Steve Levine also tried to *help* them to fix the kernel TRAP but it was not his job in this case. After the company told me that could not continue the project I went back to Steve it ook him 45 minutes to located the kernel TRAP.
The whole reason Steve never fixed the kernel TRAP was because it was the job of company from the Ukraine todo this.

The whole point being is that this management discussion with OS/2 projects is a bit more complicated then its sometimes presented in the forum.

So I can understand if you where not happy with how the project was managed overal. But it was done to best our abbilities and with the budgets we had to work. And as I always stated over the years a lot of money of sales was always reinvested in OS/2.

Roderick

[Roderick Klein]

I am working on some more details answers and will post tomorrow.

[Sigurd Fastenrath]

I am working on some more details answers and will post tomorrow.

Roderick, when I read this I fell sorry to have written what I have written because I did not want to make you so many Trouble, emotional Trouble. It was just about the Project. eCS is Long gone and it is far in the past, so please do not invest time on this Point. I know that you and others did the best they can, so - sorry, do not waste your time answering to this Point I have written and but use it for your funding. We can talk about this better, if you like, when we will met the next ime, may be in May in Berloin or December in Cologne, I am looking Forward to it.

Thanks and all the luck for the future!

[xynixme]

The problem is that generally developers have the latest machines, using a 64 OS with lots of memory and 32 bit will become more and more of an afterthought. We're seeing it in things like the recommendation to have 16GBs of virtual address space to build Mozilla as one example.

Or, less abstract and closer to home, by things like the assumption that OS/2 has a beloved Unix directory structure. I've stopped updating several apps because the old version is aimed at OS/2, while improved newer versions are aimed at the developer's eCS 2.x and AOS.

[ak120]

Or, less abstract and closer to home, by things like the assumption that OS/2 has a beloved Unix directory structure.

I don't know what's going on here? UNIX' directory structure was adopted back in the days of DOS version 2. And OS/2 remains quite compatible. Of course each has it's own limitations.

I've stopped updating several app because the old version is aimed at OS/2, while improved newer versions are aimed at the developer's eCS 2.x and AOS.

So which features you're missing?

This thread became really weird. So if there's a moderator: Please branch it out to a new thread! It has no relation with the original topic anymore. If Roderik wants to warm up old soup then I'm starting to believe in tragedies. Otčenáš

[xynixme]

when I take into account the number of bugreports in the firefox bug tracker by the main complainants I get the strong impression that complaining or even trolling is more important to some than actually helping getting bugs fixed.

Names, please. For example, people with your type of arguments tend to overlook that I'm not even really using FF for "OS/2", that I don't tend to provide personal data to create accounts to report bugs, that a bug tracker is the developer's system instead of mine, that I'm a happy user of an older version of SM, that I've never complained about Yeo's efforts regarding FF/SM/his standards, that I've guessed frame rate numbers quite frequently, that I've invested quite a lot of time in upgrades and matching downgrades, and so on. I've been critized frequently, even over here by a programmer of FF, while I'm not even using their latest non-OS/2 version of FF. I've stated frequently that efforts to keep up to date with FF's Agile are impressive. I've stated fequently that authors of ports tend to reduce the size of the OS/2 community, without recalling any serious comment that I was wrong (serious: not counting the author of FF, one of his main arguments was that RPM solved language difficulties of users of OS/2). I've stated that the author of FF for "OS/2" is allowed to release a Pentium 4 version for a 80386 OS, and so on. If I would actually use FF, then I have no reason to believe that all of my latest issues with FF for "OS/2" cannot by reproduced by a $10 Pentium III test install. The author of FF for "OS/2" has offered solutions for unreported bug over here, which just show that reporting some broken components is 100% useless: use an OS with a foreign language users may not understand (DE/EN), and buy matching new hardware. I already know that was the reply to unreported bugs, which also explains why I'm not even using nor funding their work, and I didn't need dmik's silly arguments (a.o. "I'm Russian") to presume such a "solution". And so on. Which part of "I'm not really using FF, SM is my browser" isn't always clear?

Hence the question: names (plural), please? Why are you even using a bug tracker to check people? How do you know that I wasn't a bug tracker's (virtual) "user1235"? Why are you insulting anonymous people with your troll remarks? What's your definition of a troll? I, for one, would suggest the use of a better benchmark than a system of a developer of a product I'm not using, with a CTO who first described that ignoring (potential) customers, in public, is a "great feeling" and next asks (potential) customers to send more money. An unqualified amateur. If something is broken now, then it's your benchmark. And, granted, sometimes one's initial attitude of a classic engineer.

A complicated, interesting fact is that public bug trackers do promote and support the underlying problems of commercial ICT methodes, including but not limited to Mozilla's rapid release cycle. There's no direct link, I'm not expecting an eCS 3.0 anymore. There is a reason why e.g. Agile ignores experts and why e.g. ITIL has to define that customers are satisfied. Satisfied by definition, not by solutions nor actual provided services. Users of active, public bug trackers are a part of a broken system, including but not limited to users having to donate their commercial or private data to a service provider of the author.

In a nutshell: sometimes I actually do "like" to complain about Microsoft, so according to a biased or non-evidence based point of view like yours I'm not a troll when I report everything I don't like about their supported products to Microsoft, by using Microsoft's bug tracking system with a Microsoft account, while not using or really wanting to use Microsoft's dominating products at all. Go figure...

[xynixme]

I don't know what's going on here? UNIX' directory structure was adopted back in the days of DOS version 2.

You're smarter than that and know what I mean, without having to require users to always be technically accurate. This is not an Unix forum, while perhaps appriciating a more accurate report of both Unix and the history of MS-DOS. If I want an Unix directory structure with an own root, including but not limited to solutions like RPM, then I'll start using Unix. I won't.

I could have installed such a (full) structure while smiling and not noticing it, while installing eCS 2.x DE/EN or AOS EN. But DE nor EN still isn't the prefered foreign language of the OS over here, and the developers of eCs 2.x and/or couldn't be arsed to produce an official directory structure-related upgrade for OS/2 and eCS. Different products, smaller user base. Most of the non-DE/EN community of IBM's has already left us, often without telling us.

Thanks for explsainig what Mensys did, but it's quite obvious that resources are limited. The reduced number of eCS 1.x and 2.x languages, compared to Warp 4 FixPaxk 0, has reduced the size of the community too. That's nothing but a fact. Of life. I'm not demanding all files in one directory, but I'm often pointing out that such a change can, and should be avoided. FF45 is just an example of an important product "we" aren't using anymore, and so is an updated silly Qt-based game or most of the un-OS/2'ified GCC port. If I want most of Unix, then I'll start using Unix.

[ak120]

I don't know what's going on here? UNIX' directory structure was adopted back in the days of DOS version 2.

You're smarter than that and know what I mean, without having to require users to always be technically accurate. This is not an Unix forum, while perhaps appriciating a more accurate report of both Unix and the history of MS-DOS. If I want an Unix directory structure with an own root, including but not limited to solutions like RPM, then I'll start using Unix. I won't.

Every OS/2 system has an "Unix directory structure" even when IPLed from floppy or network. And for sure every user who's posting here from an OS/2 (or NT) networked station is also using a BSD-derived IP stack. It's not a good idea to mix different topics namely directory structure, file system and UNIX. Only a simple example/question for porting trivial stuff: How about a shell script that handles "compress"ed UNIX files *.Z and also "pack"ed UNIX files *.z?

I could have installed such a (full) structure while smiling and not noticing it, while installing eCS 2.x DE/EN or AOS EN. But DE nor EN still isn't the prefered foreign language of the OS over here, and the developers of eCs 2.x and/or couldn't be arsed to produce an official directory structure-related upgrade for OS/2 and eCS. Different products, smaller user base. Most of the non-DE/EN community of IBM's has already left us, often without telling us.

For economical reasons only German counts for large remaining OS/2 deployments. There were two groups of people in the U.S. forced to learn German: DEA's dog handlers and IBM's OS/2 kernel developers. 

Thanks for explsainig what Mensys did, but it's quite obvious that resources are limited. The reduced number of eCS 1.x and 2.x languages, compared to Warp 4 FixPaxk 0, has reduced the size of the community too.

Warp 4 fixpak 0 (XR_M000) was only available in American language AFAIK. No convincing argument.

That's nothing but a fact. Of life. I'm not demanding all files in one directory, but I'm often pointing out that such a change can, and should be avoided. FF45 is just an example of an important product "we" aren't using anymore, and so is an updated silly Qt-based game or most of the un-OS/2'ified GCC port. If I want most of Unix, then I'll start using Unix.

It's not fair to blame Unix for the chaotic YUM/RPM situation under OS/2-based systems. The current partly ported RPM features (from an quite outdated release) unfortunately leads to misconceptions in this area. Unix is about small tools that make a great environment. The GNUish bloatware around is not UNIX. To make it clear I don't want to blame FSF here. And so-called open source software existed even before. There are quite good examples for portable software. But also wrong assumptions which will cause trouble under OS/2 targets sometimes torturing end-users with wrong documentation, non-working national language support or other issues. It's a effortless regurgitation to write bug reports for recurrent disturbances.

I welcome every cooperative approach to extend the coexistence of portable software. Unfortunately some people spreading rumours here seem to have their own economic interests. That's not a bad thing at all ...but there's also a category marketplace in this fora most suiteable for mountebanks.

[Martin Iturbide]

I welcome every cooperative approach to extend the coexistence of portable software. Unfortunately some people spreading rumours here seem to have their own economic interests. That's not a bad thing at all ...but there's also a category marketplace in this fora most suiteable for mountebanks.

Andreas. I only see one mountebanks that does not want to do anything and just complain that everything is a bad idea. I think you already spoke your mind and you need to move on since you don't have anything constructive to add.

Please cool off and move on. If you have a better idea just post it on some other forum thread for people to discuss it and see if it can gain adoption, funding and developers.

Regards

[Martin Iturbide]

Hi

Trying to move on there is a new post at OS2VOICE.
http://articles.os2voice.org/

It seems to be a good step on answering some of the "well funded" questions about the idea/project. I think think that Qt5 is something we need to support and that Bitwise works has shown the skill to deliver OS/2 projects on the past.

Qt's QtWebEngine seems the way to go to have a Chromium port or any other browser that uses Qt's QtWebEngine. Other ways may be too much time consuming or expensive. But if someone have alternatives it will be great to have the developers to back it up.

Regards

[Roderick Klein]

Hi

Trying to move on there is a new post at OS2VOICE.
http://articles.os2voice.org/

It seems to be a good step on answering some of the "well funded" questions about the idea/project. I think think that Qt5 is something we need to support and that Bitwise works has shown the skill to deliver OS/2 projects on the past.

Qt's QtWebEngine seems the way to go to have a Chromium port or any other browser that uses Qt's QtWebEngine. Other ways may be too much time consuming or expensive. But if someone have alternatives it will be great to have the developers to back it up.


Regards

Thanks Martin for your posting.  I just want to emphasize that the complete posting on articles.os2voice.org has been reviewed by Dmitriy Kuminov  from Bitwise works.

Roderick Klein
President OS/2 VOICE