A humble request from a new user

Hello!

I am new to this corner of the internet, but I was looking at the SCL and realized that being able to cross reference the SCL by Windows CE version would likely be much more useful than cross referencing by device. I see your other thread in here about updating the listings of software that has no device compatibility at all associated with it in the SCL, but I also see that there are others (in fact basice) where I discovered it on an archive page that claims it was made to be compatible with Cassiopeia A10/A11, but on this site nobody has made that connection yet. So in my software search for the A11, it did not appear. In this way I imagine I am missing out on a lot of software offered by this site that I would rather test and find out for myself if it works on a low RAM CE1.0/2.0 device.

Maybe I am just dumb and missed it in the drop down, but if it isn't hard to implement I, personally, would find it very useful!

Thanks for your consideration, and amazing work preserving this site. It is so so appreciated as someone just getting into my little CE devices.

Secondary thought, looking at the 24term page, I see that there are users that have confirmed it working on particular devices, yet no compatible OS's are listed. Is it hard to link compatible devices with OS versions as well? This would also vastly increase cross-reference-ability.

As a newbie I don't want to step too far out of bounds and ask you to bend backwards for me, but I suggest this as I think these could be valuable additions to searching your database!

Again, thank you for your consideration, and I appreciate your time.

At the application level. If we know about it, it's quite literally at the top of the compatibility tab, along with the CPU compatibility. The device specific compatibility info is half way down the page.

If you're referring to a filter, while the engine is capable of it, it isn't implemented because of lack of people assisting in data population. GIGO.

If you want to let us know about adjustments, please start doing so. The more contributors, the better the resource will be.

I will do what I can, my devices only support CE 1.0 and 2.0, so not sure how helpful I will be, but I will try!

Just a quick reference here:
https://www.hpcfactor.com/scl/580/Zoroyoshi/24Term/version_1.0?page=...

So if you look at "Device Specific Compatibility" here, we can see that there is evidence that this software supports:
CE 1.0
CE 2.0
CE 2.11
CE 5.0

But in "Supported Windows CE Versions"
It shows none recorded.

I am sure this can't be an isolated case. I also don't see the difference between me telling you by forum that x software is compatible with y version of CE compared to reports such as that with a device and ce version number linked together.

If you can cross-link those datasets, it might make the library somewhat more indexable.
As I see it, having OS version compatibility is more important than having specific devices to search through. This more easily allows someone (especially someone like me who is getting into this game late and has no idea what was developed for their platform) to find software that is even possibly relevant and can be tested on their specific devices. By and large, if I click on a random application in the SCL it will not typically be compatible with my OS version. At least, that's the limited experience I have so far.

Actually, being completely honest, that's what kept me on the fringe of this community for like 3 years until I looked into it again recently. Since most CE sites are long dead, and I'm late to the game, I figured my CE 1/2 device was pretty useless. I thought I needed Visual Studio 6 and extensions to program for it, I thought there was virtually no software for it, and that's what led me to the idea of gutting it to replace the innards with a RPi or other SBC (as seen on a different board). So I collected some stuff, I played with the powertoys and OEM software, and I called it a rainy day project and put it on the shelf for a while.

In any case, I recognize I'm still new and I don't want to be disrespectfully disruptive to your platform. I am just trying to offer my own personal thoughts based on my experience so far. This site has already been fantastically useful in hooking me back in and making me realize that my notions about my particular systems weren't accurate. It has pretty much solely been responsible for reinvigorating my enthusiasm.

Okay, so a couple of things from running a Windows CE site since 2000. First off, I don't disagree with you at all.

Intent
The SCL Windows CE versions fields were designed to reflect the officially states system requires from the developer/manufacturer. Most CE app development stopped after HPC2000 was released and never had any formal acknowledgement for CE .net. Thus the official spec will end at HPC2000.

When the SCL was created, that is what we - primarily the hard labours of Rich Hawley - did.

The SCL OS and Processor combinations should imply the devices that it is compatible with and vice versa. Should. Should doesn't mean it will.

When I made the HCL back at the beginnings of the site, I went down the path of CE1, CE2, CE2.11/HPC Pro and CE 3/HPC2000. There is a problem however. People don't search for "CE2.00 software". The search for "hp 320lx software". Most people don't think in those terms. There are hundreds of threads on this board and other where I've explained to people that they don't need to search for their device, that they can just find something that works on their OS and CPU. The general public doesn't/didn't "think" like that. Consequently, when I wrote the SCL, I didn't repeat the mistake. It just encourages the creation of threads and micro sites asking for/providing listings of "software for the Jornada 720".

If I'd done a better job at pooling that talent into contributing to the HCL/SCL, it would be a much better resource.

Practicalities
As I mentioned, the SCL was a labour of love to put together. Transposition of developer specs from sites/way back machine was tedious and inaccurate. No testing was done to check to see if apps worked beyond their advised spec's. Therefore where was the data to come from?
Over the years, where I've observed someone commenting that app X worked on Device/OS Y, I have updated entries. But this is very few. People haven't let us know about these in the comments, or forums or by writing in. Over the years I've granted a few others access to the SCL admin but in practice very little has been added or updated by them.

Neither either, do many people ever actually report device level success/failures to the system. There are 1074 apps in the SCL and 964 end user reports. The first application was added by me on 20th December 2012. That works out roughly at one report every 3 days... except that if I remove > My < reports from that, the total goes down to 693 user reports; one every 4 days. It isn't enough to draw relevant statistics from.

You also have to consider a) trolls and bots b) errors and c) what happens where two people make a report for device x but one says it works and the other doesn't.

Design
As I have already illuded to. The SCL engine always supported doing that mapping, in fact it was on my design spec for it in 2009 to merge down user reports and moderated reports of compatibility for OS and CPU. That version of the SCL never saw daylight and it was re-written in the 2011 version and made public in 2012. It's just that having built the SCL, unlike in the movie, they did not come. The engineering necessary to complete those modules against a data set of 693, just isn't worth the time given that we know that it will encourage spamming and trolling and in turn require additional moderation. That is the honest truth as to why it is how it is.


So if you want to make suggestions about how we get more people to contribute to the data set, or if you want to help make the data better, definitely stick around and if you want to contribute, let us know. We need people willing to add/edit and test to enhance the data set. At the current time however, given the likely return on investing the time into making the programming commitments and how tricky it is for me, I am not going to say that it'll happen any time soon.

Thank you for the kind and thoughtful response.

I understand what you are saying, and the work that you have done so far is certainly an outstanding feat. In consideration of your points I can't think of any immediate solutions but I will ruminate on this for a little while and maybe I might think of something, but I'm sure you've been down this road yourself for a lot longer than me

At my current life stage I expect that for the first little while I'll be in and out of the community, I have too many electronics hobbies (hpc, synthesizers, vintage laptops) and too little time to devote to all of them. But I do want to make an effort to work on this problem because I think CE is really worth preserving. I love the hpc platform and enjoy making my friends roll their eyes at me when I espouse how awesome it was and that I think devices should still be made to cater to the genre. But even as you just illustrated I have to keep in mind that I can't fall for the engineer's fallacy. What's right for me, isn't necessarily what's right for the masses, haha.

Anyways I'll be poking around through the SCL for the next little while at least and contributing where I can when I can. Someone's gotta do it, right?