Appreciation for TRS-80 open source projects?

[LambdaMikel]

Here is a question - how come that extremely popular projects like FreHD

https://github.com/veco/FreHDv1

- literally, hundreds of FreHDs out there! - are hugely outnumbered in the number of Github Stars by PCB replica projects such as

https://github.com/RetroStack/TRS-80-Model-I-G-E1

I am not saying that this is unjustified, but I find it interesting. IMHO, designing a novel product is of an entire different quality as designing a replica of an existing PCB.

I am interested in people's thoughts. This is not meant as a confrontation - but I am curious what causes this large dis-proportionality in attribution of impact of a product?

 

[VERAULT]

Do people care about github stars?

 

[cjs]

Well, some of us are interested more in replicas of vintage gear than in modern substitutes. So maybe that's what it's all about. And I'm not sure what you mean by a novel product being of an "entire different quality," but replicating an original PC board is a lot of work, and may well have been more work than FreHD was.

And stars do just seem kind of random, and certainly not related to quality of the product. On GitHub I have widely useful, production quality software that's maintained that doesn't get stars (dent, 1 star; pactivate, 2 stars) and large blobs of nearly random crap that does (8bitdev, 16 stars).

Do people care about github stars?

Some do, apparently. I've talked to companies who won't use my open source work because the repo has to have some minimum number of GitHub stars before they will consider it. (That's obviously fairly dumb, but there you go.)

 

[LambdaMikel]

And I'm not sure what you mean by a novel product being of an "entire different quality," but replicating an original PC board is a lot of work, and may well have been more work than FreHD was.

I guess in the sheer amount of number of components, for sure - but that's more of a "busy" kind of work that does not require any advanced (software or hardware) engineering skills or advanced understanding of the TRS-80 architecture, IMHO. That's why I am saying that novel designs are on a different level when it comes to "TRS-80 mastery" IMHO. But that's all debatable for sure! I am just thinking aloud here - again, not to be confrontational, but just for the sake of an interesting discussion of different opinions maybe.

I just feel that FreHD is maybe THE most important modern upgrade / extension for the TRS-80 today, so it seems weird that the author's ingenuity isn't appreciated more with additional stars on GitHub.

Again, I have no connection at all to the FreHD creator, nor to FreHD development, production, or sales. I just find it curious.

 

[cjs]

I guess in the sheer amount of number of components, for sure - but that's more of a "busy" kind of work that does not require any advanced (software or hardware) engineering skills or advanced understanding of the TRS-80 architecture, IMHO.

Well, I can't be totally authoritative on this since I've not done it myself, but I think you're greatly underestimating the difficulty of producing a board replica. I suppose you could just desolder every single component and socket, take high-res photographs of both sides, and then plop them into a CAD program and trace over the pictures, but I doubt that would be the normal way to do it since it's not entirely necessary and risks harm to the board. When I've done a bit of reverse-engineering on smaller boards I've always worked from a schematic (if available) and beeped things out. (Even if you do have a schematic, you still need to confirm that it actually matches the board). Regardless of how you get to your board design itself, you'll almost inevitably need to do some debugging on it, and that requires a really in-depth knowledge of the board and system.

Peripherals, on the other hand, I find a lot easier; I need only a connector pinout and an understanding of the memory and I/O maps, and I can deal with e.g. address decoding however I like.

I just feel that FreHD is maybe THE most important modern upgrade / extension for the TRS-80 today, so it seems weird that the author's ingenuity isn't appreciated more with additional stars on GitHub.

And that, too, depends on your point of view. For me, as owner of a Model I, Model III motherboard/keyboard and Model 4P, the FreHD isn't of much interest to me. I'm happy with floppies (or tape in the case of my Model I) and either the built-in serial interface on the 4P and whatever I cons up for my Model I and III (assuming I ever get that Model III motherboard running). I don't really run much in the way of third-party software on these systems; I'm mostly interested just in writing my own.

 

[AndyO]

The recreation of a vintage circuit board is an extremely complex business, and can be very time consuming. Suggesting that it's 'busy work' rather misrepresents what's involved, and how hard it can be to get it right.

That said, as the owner of a FreHD which makes my 4P a useful and usable computing tool, I also appreciate the value in supporting the vintage community with creative new products, which include things like the REX and Backpack drive for Tandy 100/102/200 models, BlueSCSI drive replacements for early Macs (and others) and any number of video adapters/converters which turn what otherwise might be museum pieces to look at into functional and useful pieces of equipment.

There's a place - and a need - for competent work in both these areas, and it seems to me there really isn't a place to compare them as to which is more or less important.

I just feel that FreHD is maybe THE most important modern upgrade / extension for the TRS-80 today, so it seems weird that the author's ingenuity isn't appreciated more with additional stars on GitHub.

I would say that however useful FreHD is, particularly in bypassing the need for working floppy drives, the most important modern device in TRS-80 world is the REX family of plugins for the Model 100 etc. This creates multiple pages of memory to vastly expand the system's original capabilities, can be filled with 'option ROM' software for extra functionality, and can make non-volatile backups. Also, even run CP/M.

Of course, it depends on which specific system(s) you're interested in. For me, my 4P is a beautifully cool piece of 40 year old computing history. It's probably the best system I have for typing on, and I use it every few weeks or so. But my Tandy 102 and 200 are used - somewhat interchangeably - almost every day, so with a REX in each and a backpack drive to move data between them and a modern Mac, I value the go-anywhere, 20 hour battery life of great portability and flexibility of use.

 

[VERAULT]

Ill just chock this up to the thread i already started a couple of years ago. The Gist being "I dont F#$king get Github".

Just throwing my hat into the mix. One thing I have noticed is how there are duplicate projects of esentially the same thing on Github.. And they are not forks or connected in anyway. Github is a mess.

 

[cjs]

Just throwing my hat into the mix. One thing I have noticed is how there are duplicate projects of esentially the same thing on Github.. And they are not forks or connected in anyway. Github is a mess.

Well, you would really hate this thing called the "Internet," then which is even worse. Anybody can go and create their own site there, and put whatever they want on it, and there's duplication of not just "essentially" but exactly the same thing all over the place. Look for something as simple as a CP/M 2.2 manual, and there are copies on dozens of different sites, not even all the same scans, some HTML versions of the original manuals, even. It's crazy what happens when you don't have a central authority dictating who can post what where.

 

[VERAULT]

Well, you would really hate this thing called the "Internet," then which is even worse.

No not the same thing at all. Keeping on OP's topic

 

[VERAULT]

I guess in the sheer amount of number of components, for sure - but that's more of a "busy" kind of work that does not require any advanced (software or hardware) engineering skills or advanced understanding of the TRS-80 architecture, IMHO. That's why I am saying that novel designs are on a different level when it comes to "TRS-80 mastery" IMHO. But that's all debatable for sure! I am just thinking aloud here - again, not to be confrontational, but just for the sake of an interesting discussion of different opinions maybe.

I just feel that FreHD is maybe THE most important modern upgrade / extension for the TRS-80 today, so it seems weird that the author's ingenuity isn't appreciated more with additional stars on GitHub.

Again, I have no connection at all to the FreHD creator, nor to FreHD development, production, or sales. I just find it curious.

I agree with you about the FreHD but the problem can also be group related. Where is the Central repository for all things TRS-80, for that matter the same question for Apple II, ATARI, or the COLECO ADAM. Depending on the question It may be an easy answer or draw a total blank. And the size of the group (fans, collectors, hobbyists, etc) varies drastically by model or brand... I feel like enough is enough on the Commodore 64 scene at this point, however other machines get no attention whatsoever.

 

[cjs]

No not the same thing at all. Keeping on OP's topic

The OP's topic was not, "what a mess GitHub is," but "why aren't certain things I like ranked as less popular by people voting on the Internet"? I don't see that your complaints about GitHub (or the Internet) letting individuals post whatever they like rather than having some central authority determining what and how people should post is related to this at all.

And your complaints, that amount essentially to "You folks writing software and sharing it with others are doing it all wrong," are not appreciated. If you wrote open source stuff and actually collaborated with others on it, you'd appreciate GitHub and similar things. But instead you complain about something you know nothing about, and those who are contributing to the community.

Where is the Central repository for all things TRS-80, for that matter the same question for Apple II, ATARI, or the COLECO ADAM.

There isn't one, for the very reason that there was never One Central Repository for all things TRS-80/Apple II/etc. back in the days when we put this stuff out on paper and floppy diskettes.

And even beyond that we allow individuals to put together whatever pages and resources they like on the web, without having to submit to a central authority to make sure they're doing it the Right Way, the very nature of different materials wants different kinds of sites for them. For archival copies of software and documents, archive.org is great. But it's terrible for having discussions: for that a forum site is better. Both are terrible for distributing and managing software under development, which is why we have GitHub, GitLab and similar sites.

But let's look at stripped down version of one of your complaints: discussion about the Apple II. Without any thought at all I can come up with four different discussion sites that cover the Apple II: applefritter.com, forum.6502.org, retrocomputingforum.com, and this site. (There are several more, and for simplicity I'm leaving out several Discord servers.) Now which ones of these do you want to shut down (or shut down their Apple-II-related conversations), and how do you intend to convince them to do that?

 

[VERAULT]

There isn't one, for the very reason that there was never One Central Repository for all things TRS-80/Apple II/etc. back in the days when we put this stuff out on paper and floppy diskettes.

It was rhetorical. I know there isn't one. I was just making the point to @LambdaMikel that the scattered nature of the "scene" is part of the problem.

 

[cjs]

I was just making the point to @LambdaMikel that the scattered nature of the "scene" is part of the problem.

Ah. There certainly is a reasonable possibility that the group voting on GitHub with stars is not representative of the overall TRS-80 retro-community but is skewed more towards "developers" of some flavour or other, and that group tends to prefer replica PCBs to modern substitutes for older storage systems.

I don't see how that requires or even relates in any way to a rant about GitHub, though.

 

[VERAULT]

I don't see how that requires or even relates in any way to a rant about GitHub, though.

Who is ranting? The OP just asked a question.

 

[LambdaMikel]

Thanks all for the discussion. Correct, I don't intend to rant about anything - these are just my humble opinions and it is not meant as confrontation or something you should feel a need to comment on, so please don't get angry with me Everybody is entitled to an opinion, and this is mine. And I am interested in your opinion, hence the thread here

1. IMHO, GitHub stars do matter - not everybody is retired, and having projects on GitHub with many stars DOES make a difference for job hunting (not sure the FreHD author is retired or not, I don't know anything about him! sure, his repo is a bit underwhelming in terms of presentation - other projects do MUCH better there. However, that is also largely irrelevant IMHO given the huge success and importance of FreHD).
2. GitHub stars are the only currency and reward that developers get in return for their efforts of sharing their hard work with the community, for free. We want to make sure that it stays like that - so reward them if you can.
3. Frequently, others than the original developers benefit from their hard work.
4. IMHO, everybody that gets a copy of such a device from a manufacturer / shop / reseller should also consider leaving a star on GitHub if the device is found to be useful. It's the least we can do for the original developers.
5. I know that making a PCB replica can be as easy as making a scan and tracing the connections (done it myself with a keyboard membrane). This was also done with countless retro computer boards - recently, for the Amstrad CPC 464. This resulted in problematic connections that then gives YouTubers plenty of material to make "content" from (
   ). Sometimes, entire new boards with doubtful features such as vacuum tubes for audio (EVO64 C64) are made, or the Model 1 replica above, which argubly requires much more skills to make for sure.
6. In case of the Model 1, the schematics are available - sure, there might be some bugs in the schematics, but you always have the original available and can easily recover from any inaccuracies in the specs simply by looking at the working original. In contrast, developers of new hardware, original designs, don't have an original... it doesn't exist.
7. This is not meant to turn into a flame war - as I said, I appreciate ALL open source projects for the TRS-80, and frequently leave stars even for projects that I don't use (like the Model 1 replica board). And I encourage you to do the same.

Cheers!

 

[LambdaMikel]

Peripherals, on the other hand, I find a lot easier; I need only a connector pinout and an understanding of the memory and I/O maps, and I can deal with e.g. address decoding however I like.

That's such an interesting statement... as if designing the IO interface would have anything to do with the complexity of operation that the peripheral performs (also, it needs drivers and what have you!) It's like saying making a car is easy cause you only need to get the steering wheel right.... anyhow. Just my 2 cents.

 

[Eudimorphodon]

I've already said this about the VIC-20 and C64 replica boards but... I guess my befuddlement remains. Is worn out PCB material really that widespread of a problem if you're trying to restore one of these machines?

Personally I guess I'm of the mindset if you're going to bother making a complete PCB to replace the board in an original machine that just seems like it's too much of a basket case to repair(*) you might as well update it to use more readily available parts if the original uses genuinely obsolete and hard to find bits. Like in the case of a TRS-80 you might as well use a generic ROM instead of the mostly unobtanium character generator, and maybe some more modern SRAM for the video memory. (The Model I also has a pretty darn fiddly power supply section that Radio Shack only did that way because the pile of loose bits they made it out of was slightly cheaper than the newer LM7805-style regulators.) Unless you're actively trying to fool someone into thinking your machine is an original it kind of feels like cosplaying specifically to make your life harder, and the result is a less reliable machine.

(* Although, really... unless you've snapped the board in half there's probably little you could do to an original TRS-80 that can't be fixed with bodge wires; most TRS-80s that were actually used are pretty much made of bodge wires so I'd say more hacks and bad soldering just enhances the original patina.)

 

[LambdaMikel]

I really hope that nobody is using "irreparable " (such a thing does not exist! ) M1 boards and scrapes them for chips to be plugged into replica boards... after all, we should try to preserve old hardware. I personally never came around a retro computer whose board was the problem, really. So it must be that there a dozens of people out there building M1 replica based on the number of GitHub stars? What's the practical use of an M1 replica board otherwise?

 

[Eudimorphodon]

I personally never came around a retro computer whose board was the problem, really.

I repaired a Commodore PET that had a surprising number of rotted out traces on it, but fixing a rotted trace isn't really rocket science. It's not like Model Is had onboard batteries to melt holes in them, and even if they did, well, it's only a 2-layer board so, again, just bodge it.

 

[VERAULT]

Macintosh boards (Macintosh II in my particular experience) succumb to battery and leaky cap damage that had rendered MANY, MANY of them un-repearable.