Cloverleaf Campaign is Live

I wonder who wrote the following:
“IMO, in a market as small as RISC OS’, fragmentation into different OS distributions is as silly now as it was back when we had the ROL/Castle stuff.”

Dunno, “give us a clue”

Perhaps, the statement above suggests the motive behind attempts to publicly trash the Cloverleaf Kickstarter.

?? Andrew R has been supportive, here and behind the scenes from all accounts.

The general trend in stuff above is that this time the failure seems to be effective communication and a misaligned sales pitch. I covered ways in which I think it is wrong and what I believe would be better. (Not “best” because I’m a firm believer that you can always find improvements)

Daniel is of the opinion that the content is not something that should be a kickstarter, much of his reasoning seems to align with the idea that he didn’t understand what the offer was, and that comes back to my point. Reinforcing that is Mikko with “It’s just a shame the Kickstarter projects have been framed so poorly.”

Wading through the most recent kickstarter is something I haven’t done. A skim read tells me it isn’t succinct enough.

So can you please answer me on which part of these OS you have proof they are different in the source/API/coding style? I am not aware of any substantial difference, so I’d like to learn what makes you concerned in their code bases differences (if any of course).

I never claimed they were different in terms of source, APIs or anything at the coding level.

The point was, with so few users (comparitively to other OSes), is there really a need for multiple distributions with different bundles of included stuff? Surely concentrating efforts into one main release would make far more sense. Less confusing, and less duplication of effort to create/distribute/support them and everyone gets the “best” distribution.

It also works against the OS when trying to attract new users. New users who aren’t familiar with the OS wouldn’t have a clue what the differences are or how they should choose between them. Given that the official Raspberry Pi SD card imager includes the ROOL distribution, surely it’d make far more sense to focus efforts on improving that rather than creating other competing distributions since that’s where newer users are more likely to try the OS out.

Reinforcing that is Mikko with “It’s just a shame the Kickstarter projects have been framed so poorly.”

I said that and stand by it but I also second my support for Cloverleaf’s efforts: I think they’re a positive input to RISC OS.

I am sorry Andrew… you are being confusing now (not joking or being sarcastic)

I never claimed they were different in terms of source, APIs or anything at the coding level.

You say you’ve never claimed they were different, but you compared Cloverleaf to the ROL/Castle situation where they WERE different…

The point was, with so few users (comparitively to other OSes), is there really a need for multiple distributions with different bundles of included stuff?

Which OSes? ROL/Castle didn’t have much bundles difference (they both came with the original Acorn bundled software), they were actually different OSes… (and still are, hopefully less in the future).

So, if I understand you correctly, you meant the differences in terms of bundled software not the OS, am I right?

But the bundled software is something that ALL vendors have different: For instance RISCOSBits software bundle is different than the software bundle on an RComp ARMBook. And btw the software bundle on an ARMBook is also different than the software bundle on a mini.m that is sold by the same company.

Vendors even try to customise some of the RISC OS main icons! So almost none of them has the RISC OS cog (I believe only RISCOSBits), no wonder we are all confused on which one is the actual RISC OS logo :D

Surely concentrating efforts into one main release would make far more sense. Less confusing, and less duplication of effort to create/distribute/support them and everyone gets the “best” distribution.

But isn’t the software bundle something to incentive the sales? And, honestly, but even ROOL have the Vanilla RISC OS and then the ePic RISC OS which is paid and comes with bundled software…

So why are you pointing your finger at Cloverleaf for this? What about RComp? RISCOSBits? ROOL??? (I think the only one who leaves RO 5 as Vanialla RO5 is CJE!)

I think it’s perfectly normal to have different bundles, but the different cog IS confusing I am with you on that one (even if you haven’t mentioned it!). The most confusing one is actually RO5, that uses the ROOL cog as the RISC OS logo… or is it also the RO 5 logo? wait I am confused again… damn! XD

Hope I haven’t been too aggressive in my response, if I have please accept my apologies.

Given that the official Raspberry Pi SD card imager includes the ROOL distribution, surely itâ€−d make far more sense to focus efforts on improving that rather than creating other distributions.

There is exactly the flaw in your approach. The Cloverleave distribution is also based on the ROOL distribution because that is the basic operating system wirh the last improvements. There are differences in the bundled software and in the added “Clowerleaf tools and drivers”.
As Stefan wrote, everything that he can influence will be “open source”, so if it brings improvements, it will probably also be incorporated into the ROOL distribution.
And the Kickstarter wants to attract new users. A good bundle makes a lot of sense.
As an “old user” I have licenses for all programs I use and I know how to create RISC OS media that meet my needs.
Another problem is the same that I have. English is not my main language. Misunderstandings are likely. ;-)

Seems like something has overlapped and Paolo has similar arguments. ;-)

It appears pretty simple to me, the previous Kickstarter would be at its best be described as ‘ambitious’, had a target price of 50,000 euros and a public admission from Stefan that 100,000 euros would be required to meet the project’s goals.

This second Kickstarter seems to have even more work needed in it, and yet a target price of only 6,000 euros.

There’s a lot of giving people the benefit of the doubt, “never wanting any negative publicity or discussion” and “never wanting to upset someone promising the moon” in these here RISC OS circles, but surely people are allowed to notice the obvious discrepancies here and share their concerns?

There has been many times people have been burnt by over-promises and under-delivery and perhaps a more cautious approach is needed by punters.

There are several people in the thread who correctly state “It’s not my job to tell others how to spend their money”, which I agree with. I can also state unequivocally that “I’m not pledging any money towards this”. Make up your own mind.

PS. The previous Kickstarter campaign mentioned me by name. I would like to make it clear that I have absolutely nothing to do with the Cloverleaf Project or Kickstarters.

So, if I understand you correctly, you meant the differences in terms of bundled software not the OS, am I right?

Not really, you’re reading far too deeply into it. I was merely using it as an example of two versions of the OS that, to people who don’t already know anything about RISC OS, are basically two versions of the same thing. Yes, ROL/Castle’s OSes were different in their approaches and internals, but as an outsider without the back story and history of that, you’d have no idea because they look and behave the same and wouldn’t have a clue how to choose which one you should go for.

I also never made any claims to have thought through all the details of how a single distribution would work – because I haven’t. I merely stated that in my opinion, having multiple distributions of effectively the same thing with different bundled extras in such a small market is silly.

At the end of the day, the comment on fragmentation was a very small part of the problems I see with Cloverleaf and their approach and was, as I say, just my opinion. You’re free to disagree, of course, but look at it from the point of view of someone who’s never used RISC OS before and sees it in the Raspberry Pi imager and decides to try it. Would you rather thay get a version that’s just the plain OS without much in the way of extras to play with, or would you rather the various companies producing these distributions work together and make the most publicised and easy to get distribution a better example of what the OS can offer by including the best bits of all of them in the one distribution that people are most likely to see and try out as their entry-point to the RISC OS world?

Yes, there’s flaws in the idea, it’d be a bit of a pain licensing-wise and sure, there’d still need to be room for commercial upgrades to be available later. As I say, I’ve not thought through every little detail of it. That wasn’t my intention.

By far the bigger issue with Cloverleaf is that it’s promising to do a lot of things that it’d requrie a lot of money to do, but it seems to be completely unable to show how it would do those things, specially with the newly revised goal of only 6,000 Euro. Both Kickstarters have been completely devoid of any plan for how they’re going to spend the money they may or may not get or how they’re going to go about developing the rewards they’re promising. They want you to send them money on the blind hope that they know what they’re doing when everything they’ve shown so far suggests they clearly don’t.

Part of the guidelines for honest and clear presentation on Kickstarter say:

(Do not) Set a funding goal that won’t allow you to bring your project to completion, or accept funds on Kickstarter with the knowledge that you’ll require future outside funds in order to fulfill all of your rewards.

As Peter notes, Stefan admitted during the first kickstarter that he’d need at least 100,000 Euro to complete his project, yet only asked for 50,000. Now he wants us to believe he can do it for 6,000. Again, though, there’s absolutely no plan listed with the Kickstarter on how he plans to achieve that or how he’s planning to spend the money he gets. All you get on the Kickstarter page is a “Please trust us” but there’s nothing to show why we should trust them. And, of course, when you ask for any clarification the standard procedure appears to be that you get shouted at and blocked from their channels for daring to question the project. That’s a great way of earning trust, right?

(As a side note, when I quoted that page on TIB, someone pointed out that they’re not rules but guidelines. I’m aware of that (and was when I made the original TIB post), but I’m pretty certain there’s a reason Kickstarter provide those guidelines and it’s precisely to avoid situations like the Cloverleaf mess. If Cloverleaf actually followed all the guidelines and the project was presented clearly and transparently with a full plan, we’d probably not be having this discussion.)

PS. The previous Kickstarter campaign mentioned me by name. I would like to make it clear that I have absolutely nothing to do with the Cloverleaf Project or Kickstarters.

And how many other people who Cloverleaf have said are working on things that their project will deliver are not actually involved in the project? Stefan keeps throwing people’s names out as if they’re part of his project, but as far as I can tell, almost all of them have nothing to do with it. Again, there’s no transparency here at all and Stefan wants you to send him money for things that he’s making out to be part of his project but are actually entirely separate and in most cases are already happening anyway with or without funding from Cloverleaf. Those things are far better suited to something like ROOL’s bounties than a Kickstarter campaign.

One really does wonder what actually came out of that 8,000 Euro they spent on a “crowdfunding specialist”. Doesn’t seem like money well spent as far as I can see – specially when the current Kickstarter is asking for less money than they spent on that specialist. No matter how I think about it, I can’t make that make any sense whatsoever.

Of course, decide for yourself if you want to give your money to the project. It’s your money and I don’t care what you do with it. Personally, I think there’s so many red flags in this Kickstarter that I do not see any reason to support it. Every time I look at the project, I discover more things that’re concerning about it and the complete lack of transparancy certainly doesn’t help.

If Cloverleaf actually followed all the guidelines and the project was presented clearly and transparently with a full plan, we’d probably not be having this discussion.

Totally agree.
My suggestions were a nudge towards that clear statement.

One really does wonder what actually came out of that 8,000 Euro they spent on a “crowdfunding specialist”. Doesn’t seem like money well spent as far as I can see

Oh, I hadn’t noted the money spent on that. If Stefan had advice, was it followed, or is the specialist not so special?

Every time I look at the project, I discover more things that’re concerning about it and the complete lack of transparancy transparency certainly doesn’t help.

I suspect that is a native English speaker dealing with text from a non-native source; plus a bit of Uber-enthusiasm causing Stefan to be unfocussed and missing delivering the message.

Enough on this, Stefan will feel we are kicking him whereas the intent of all here, I think, is to point a better direction.

I suspect that is a native English speaker dealing with text from a non-native source

Possibly, although just for the record I absolutely understand Stefan’s first language isn’t English and that’s exactly why I’ve not made comment on the spelling and grammar side of things and have only commented on the project itself.

(And yes, I didn’t spot my own typo in “transparency” when proofreading my message. :) )

This tends to remind me of running ‘Chocks Away Demo’

Of trying to practice landing – when some pesky Jerry tries to shoot me up!

(Wrong way around – but I think I give the idea)

Oh, I hadn’t noted the money spent on that.

Did you notice the €7,000 spent on advertising on Facebook? Between that and the overpaid “specialist” (seriously, that’s like six months of work at what I’m paid), we’re already up to €15,000. Is this money being well spent?

As Stefan wrote, everything that he can influence will be “open source”, so if it brings improvements, it will probably also be incorporated into the ROOL distribution.

The problem is that, with one exception, I’ve never really seen the planned licensing terms for Cloverleaf projects made clear.

On the subject of that one exception, the intent was clearly not Open Source. When we were discussing the IDE, the proposal appeared to be to take existing projects, embrace and extend them, then release the results commercially to make money. The interest that Stefan had in WinEd appeared to evaporate when it was explained that its use of the GPLv2 licence might make that a legally tricky approach to take – although I did get an email later on telling me that the icons were nice and asking if I designed them¹.

Some absolute clarity, goal by goal, on what licences will be applied to the individual deliverables would go a long way to reassuring me about Cloverleaf. Ideally the exact licences to be used, and not just a woolly “open source”, because that can be significant too – especially for things to be integrated into the OS down the line.

¹ No, I didn’t. There’s no credit that I can see, but they’re quite distinctive in style so I could probably hazard a guess as to who did and I’m sure that others could do so too.

Did you notice the €7,000 spent on advertising on Facebook? Between that and the overpaid “specialist” (seriously, that’s like six months of work at what I’m paid), we’re already up to €15,000. Is this money being well spent?

The social media stuff is odd, to say the least. I don’t know where the money is going, but it’s not on a human to personalise the outreach. I got a email a while back congratulating me on my RISC OS-related blog which apparently the writer read(!) and asking if I would consider joining their Facebook group.

If anyone can tell me where my blog is, I’d be very grateful – I seem to have inadvertently mislaid it, and it’s not where Cloverleaf told me it was.

There’s no credit that I can see

It was me :) If you need any updates, just let me know!

The social media stuff is odd, to say the least.

I can’t say I pay any attention to Instant-Twit-Face. I was going to say “much attention” and then I realised that pretty much the only thing I know about that stuff is:

1. What the wife reads out at times, is sometimes amusing
2. The names of some of the services¹ integrated above.
3. The content seems to be vapid at one end and extremist at the other, how much sits in the middle I don’t know.

As a platform for adverts, well, lets just say I employ filters on browsers specifically to stop adverts and particularly the hyper-active graphic ones so if I did do that SM stuff I’d be blocking the ads.

¹ Those I’ve probably come to references of regularly since the 2020 election in the USA, where apparently these services were wonderful (Trump and supporters using them to apparently spread disinformation) and the most evil thing on the planet (Trump and supporters when the disinformation distribution was stopped)

Okay, having read this thread for the past few days, I’ve decided to wade into the mire. Please do not take this as in any way “bashing” Cloverleaf, it is intended as constructive criticism.

What is actually being offered?

Firstly, what the hell do you actually get? Or, let’s put it differently. Let’s assume the target is reached. Given the previous project managed to raise about half of the €50K being asked for, I do not imagine €6K to be difficult. We’re already over halfway there at time of writing.
So, the funding is secured. Now let’s assume that I give my €49 for the ArtCube photo editor, and I’m the only person to have done so (it currently has 0 backers). Will I get ArtCube, or not? If not, what would that money be used for?

How does anybody on earth plan to improve RPCEmu, port RISC OS to a new platform, give us WiFi, extend ChatCube, create a photo editor, etc etc for less money than was spent on a so-called specialist? It’s completely unrealistic.
To put this into perspective, look at https://www.riscosopen.org/bounty/

Would these pledges/rewards not make more sense to be done in parts? Estimate how much it might cost to create a photo editor or provide WiFi support, and then set up a funding for exactly and precisely that? In that way, people will know exactly what they are giving money to, and exactly what they are supporting, and exactly what they should be expecting to see as a result of their donations.
Nobody wants a Vega+, so things should be both realistic and clear.

Language

I don’t want to pick on the use of language, as I very much doubt that I could write anything like that in French, a language I’m supposed to know. Have a look at https://riscos.fr/frob31/ and decide for yourselves. ;-)
Suffice to say, given that we’re mostly an anglophone community, it would be a good idea to get a native English speaker on-board to go over the draft documents before they are published.
Because every language has its odd little rules, and we English speakers all know insinctively that it’s a big yellow bus that takes kids to school, not a yellow big bus (or for the Frenchies, a big bus yellow ;-) ). Having things a little muddled like that isn’t a good image when it’s something that’s not actually that hard to fix by simply asking for assistance from a native English speaker.

Please note, I could say the exact same thing about the company that I work for. Some of their translations are… a little odd in places. I did offer to help years ago, but like a lot of French beaurocracy, having the requisite qualifications counts for so much more than having actual experience. I can offer something that not one single Frenchie-that-went-to-uni can offer, and that is 42 years (at the time) of speaking English as my native language.
But I don’t have a degree so it doesn’t count. <shrug>

Numerous factual errors

First bullet point: Both have been developed together by Acorn Computers Ltd. in Cambridge, UK in the 1990’s.

First paragraph after the initial bullet points: Sadly Acorn decided to close down their computer division in 1990 and only continue support the ARM technology.

Both are wrong.

The ARM processor was born around 1985-1986, and the first machine with an ARM processor was sold, along with Arthur, in the summer of 1987.
If you recall Paul Fellows’ talk on the failure of ARX and the shush-bang-ali they did with the multiple copies of the clock app… remember that all of this happened before the first Archimedes was released.

I’m not entirely sure when Acorn threw in the towel, I think it was 1998 but I can’t be bothered to look it up. Certainly, if Acorn gave up in 1990, then where did the RiscPC come from? Or the A5000 for that matter. ;-)

But not only ARM chips are amazing but also its original operating system RISC OS.

Reads like a fanboy. I don’t want to know what you think of the processor and the operating system, I want to know why I should feel the same way.
Or, to put it differently, what can RISC OS on an ARM offer me that a mid-range Android tablet, or an equivalent priced small x86 laptop cannot?

far better than Windows 3.1 or Windows 95 – subjective. While there were features of RISC OS that were indeed better than both of those, they in turn had features that were better than RISC OS. Besides, one of the main criticisms of Windows was the complete lack of coherent memory protection making General Protection Failure a common occurance, and let’s not talk about how often W95 would puke up it’s infamous blue screen of terminal distress. It happened so much that there are even tropes named after it, when applied to people! https://tvtropes.org/pmwiki/pmwiki.php/Main/HeroicBSOD
Microsoft got their crap together. You can still BSOD XP and such if you have shonky badly written drivers running with privileges, but errant malfunctioning programs generally don’t take the system down any more.
RISC OS? I could trash your system in three instructions. :-)

Do you remember when I was trying to build ARCbbs and random bits of everything started to die? That’s because an error in calculating offsets meant that bits of data were being spewed over top of other module’s code and workspace.
If RISC OS had a mechanism for noticing and reporting a BSOD, it would have. Instead, some stuff fell over and eventually it just stopped responding.

Since then RISC OS was mainly supported by enthusiasts – it’s rather fascinating how this story completely omits the work done by ROLtd, and the huge amount of work Gerph put into shoring up the shonkiness of the OS to improve both stability and being able to work out what went wrong when it went wrong. All features lacking in the current version of the OS.
How often have we said “do *MemoryI PC-20 +40” or “what does *Where say?” and so on. Hell, even the nicely detailed Debugger exception dumps curiously omit to provide a disassembly around the point where things went tits up.
Gerph, on the other hand… damn, just read this: http://gerph.org/riscos/ramble/testingdebugging-2.html#BTSBTSDumpDiagnosticDump

It has integrated BBC BASIC with access to ARM assembler. – I don’t think that’s really been a selling point for the past two decades.

Also available are GCC and Python. – and Lua?

as the access to the ARM GPU is blocked by ARM Ltd. as they only give programming information to license holders. – is this correct? I know ARM designs GPUs, but don’t TI and Broadcom as well? Yes, the world of GPUs is wrapped in NDAs. Which is why a lot of it is accessed by using special binary-blob drivers which have a fairly generic interface (think DirectX or OpenGL). We lack that.

But, even if Broadcom decided to make a big community effort and release the full specification of the GPU in the Pi family… you know what? It would make about zero difference.
The days of somebody writing a good game in their bedroom has long passed. Take a look at the level of detail expected of a modern game (such as The Last Of Us), the quality of the rendering, the number of objects and actions and interactions. Entire teams of developers create those sorts of things. There were probably more people working on that one game than have ever worked on our entire OS, ever.
I would omit this entire paragraph, and just go straight to promoting retro gaming.

Clarity desperately required

!Iris will be released in a few weeks. – it will?

The main goal of this Kickstarter and our project is to speed up the development for RISC OS. This includes programming missing hardware drivers like WIFI, Bluetooth, USB 3 and NVMe mass storage. Another topic is filling gaps of missing software or improving the quality of RISC OS apps to current standards. – how is this going to be achieved for €6,000?

(beat)

We cannot take care of all that with just this Kickstarter project – and here is where the questions start, because one statement contradicts the other. You have a Kickstarter project to do X, Y, and Z…but you cannot do X, Y, and Z with just the one Kickstarter project? So what are you doing in this Kickstarter project. Note the words in bold. A person is not going to be interested in grand visions, if you want people to part with cash, there should be specific “this project means this will be done”. Nothing more, nothing less.

As has been mentioned by others, I don’t see any clear indication of what this project aims to actually achieve. Because all of the rewards on offer? Asides from the T-shirt, I cannot imagine how you’re going to get any of it done for €6K, never mind all of the things listed in the Rewards.

A little further down: We also don‘t expect to receive a full funding for the packages we offer.
This would be a red flag to many. People like things to be kept simple. “You think you can do X for €Y. I believe in you so here’s some €€ so you can work on making X happen.” That’s how it usually goes (Planet Gemini), or not… (Vega+).
There’s a clear objective. A clear goal. And people donate to make it happen.

It seems to me that Cloverleaf is lacking the clarity. I get Stefan’s enthusiasm, but perhaps the desire to fix everything needs to be toned down to tackling things in turn. A funding project for WiFi, for instance. And a different one for ArtCube. And another for GPU support… and so on. That way, things are both clear for the inverstors, and also for you. You’ll know what people think is important (it makes the level of funding being requested) and investors will know that you’ll be concentrating on delivering that thing.

Final thought

RISC OS has languished for many years. It won’t hurt to attack the deficiencies one at a time rather than all together.

and the most evil thing on the planet

Authentic Steve, please read the “Oh FFS” part of https://heyrick.eu/blog/index.php?diary=20210115 (about halfway down).

There are a number of people at work who think the same sorts of things, because guess where they get their “facts” from… :-/

totally agree:

I get Stefan’s enthusiasm, but perhaps the desire to fix everything needs to be toned down to tackling things in turn.

and

Final thought

RISC OS has languished for many years. It won’t hurt to attack the deficiencies one at a time rather than all together.

amen and I would add that changing things IN the OS requires coordination as well.

far better than Windows 3.1 or Windows 95 – subjective.

Hmmm, 3.0 was total poo. 3.1 was almost usable (3.11 was usable)

Win95 looked like a glitzed up RO, but RO never had a pretty blue crash screen – except as a screen saver IIRC

Iris will be released in a few weeks. – it will?

It’s sort of half out, with people who purchased OBrowser, replay the talk Andrew Rawnsley did for the Wakefield show.

Like there’s a new TCP stack, because R-Comp/ROD needed it for a customer. IPv6 is in there, WiFi is a small enhancement apparently, and it works with a few NICs. It just needs appropriate drivers to match interfaces other than the ones R-Comp are using.

It seems to me that Cloverleaf is lacking the clarity. I get Stefan’s enthusiasm, but perhaps the desire to fix everything needs to be toned down to tackling things in turn.

Distracted by “shiny” in mirror factory…

It’s sort of half out, with people who purchased OBrowser

It’s available now; you get access to Iris by purchasing OBrowser from the Store. I did, and I’m happy to say it was worth it. At some point, it’ll be free.

replay the talk Andrew Rawnsley did for the Wakefield show.

A good idea! To watch the video of the talk.

Rick mentioned the Vega+, after that debacle every tech journalist on the planet is scanning Kick-starter projects for anything that over-promises and under-delivers – RISC OS doesn’t need that sort of publicity, the shit will stick to everyone.

In German – what defines thirty thou?

ie 30.000 or 30,000

Rick mentioned the Vega+.

That did so much damage to the whole retro-community. Utterly unbelievable what went on there.

just go straight to promoting retro gaming.

Thankfully Cloverleaf has not gone the same route as the first and just assume he could pirate retro software. I raised the issue with regards to copyright/permissions on 10th December last year in this forum (Link: https://www.riscosopen.org/forum/forums/1/topics/15883?page=6#posts-112105 along with other questions) but was just completely ignored. I wasn’t subjected to the same level of hatred as Andrew Poole received when he later raised it on Facebook.

It’s interesting to compare ROOL’s Bounties with Cloverleaf’s Kickstarter. Bounties allow us to contribute money in the hope (and nothing more) of getting what we want. Kickstarters seem to be rather more rigidly defined – either they succeed or they fail.

Maybe if we got less hung up about this binary success/failure result, and regarded Cloverleaf’s efforts as simply an alternative scheme to provide a financial inducement for some RISC OS development, we could all chill a bit.

But I don’t know if Kickstarter even permits that.