open source model

Community Forums/Monkey2 Talk/open source model

skid(Posted 2015) [#1]
I still can't get my head around open source.

If patreon earned a place on the test team, BRL published a generous free version as opposed to a crippled demo version, the source code could remain proprietary.

I loathe the way bmx3d.googlecode sunk like a stone.

Anyway, to encourage an open minded reception to my idea I'm off to double my patronage. Go mk2!


Pharmhaus(Posted 2015) [#2]

Maybe it's not a good idea?


It is a good idea. I hate it that I don' t have access to bmk.
The only thing which could be improved is the interface e.g. contribution guidelines and such beside that it is great.


I loathe the way b3d2 sunk like a stone.


I think this is a common problem of engineer type of people which only think of the idea but ignore everything else.
Improving the documentation would be good starting point. UML Diagrams, comments, known bugs and ToDOs don't hurt neither.
It is simply naive to believe that someone else gets into your code mess base with ease when none of these exists.
Changing things or adding features is hard and why would you if you can simply buy a well documented thingy for low budged.


GW_(Posted 2015) [#3]
I loathe the way b3d2 sunk like a stone.

You mean the Max3d module? That thing is awesome! Maybe we should start a patreon to have someone [smarter than me] flesh it out.
I'm fine with M2 being closed source. but the build tools would need to be OS and there needs open ways to look into the compiler. for example to access to the ast-tree etc.


MikeHart(Posted 2015) [#4]
I am totally against closed source for M2. If it becomes closed source, there are other closed source solutions that provide a bigger and more complete feature set.


Xaron(Posted 2015) [#5]
Could you name that one Mike?


MikeHart(Posted 2015) [#6]
Corona, AGK2, GM:S, Unity, whatever. So many...


Danilo(Posted 2015) [#7]
Just a reminder:
It was Mark himself who decided to make MX2 completely free and open source, and he introduced Patreon to many of us.

http://marksibly.blogspot.de/2015/05/monkey2.html
Saturday, May 2, 2015

Monkey2!

Ok, some fairly big news: I've decided to start work on Monkey2!
I'm going with an open source/crowd funded approach this time, as I don't consider the
idea of selling languages to be commercially viable these days. I am planning to start
with a patreon project (see: https://www.patreon.com/) which I hope to have up in a
week or so.

https://www.patreon.com/monkey2

I can't see a reason to change that now, and break one's word.


Richard Betson(Posted 2015) [#8]
Really I feel Mark should promote his Patroen page more prominently on the website. It's a good idea and plan. One I agree with.

Why not promote his Patreon page in your forum signature. :)


Danilo(Posted 2015) [#9]
Promoting the Patreon page probably doesn't help much, as long as there is no useable product/project to play with -
including libraries, documentation, nice IDE. I think at least outsiders and newcomers probably wait until there actually
is something to test (and to support if they like it).
Interested people can't test MX2 freely now, and it's probably better to wait with public release until it's pretty complete (incl. docs),
and just working out-of-the-box. A non-complete or non-working/buggy thing could scare interested people/testers away after first look.


nullterm(Posted 2015) [#10]
Indeed. I'm working on a "try to sell the product while we haven't even finished it yet" and it's a gong show (don't look at me, biz guy can't help but say "yes" to people with money). You get stuck in this cycle of "well, it's not done yet..." "well what do we need to add?!?" ... then cram a ton of features into a unfinished product. When we should kick out a smaller, but well thought out and tested product. It's easier to avoid that cycle by waiting until the core (aka Minimum Viable Product) is solid and fully baked.

I think Mark is going the right route, small snippets every now and then just to get feedback. This isn't Mark's first rodeo, so I'm confident and excited to see where it's headed.


JoshKlint(Posted 2015) [#11]
Typically, free and open source signals a lack of confidence, and goes together with a poorly defined feature set, lack of documentation, and bad user experience. The attitude tends to be "you've got the source, you fix it". As a general thing.


Nobuyuki(Posted 2015) [#12]
Typically, free and open source signals a lack of confidence, and goes together with a poorly defined feature set, lack of documentation, and bad user experience. The attitude tends to be "you've got the source, you fix it". As a general thing.


a half-assed commercial product is even worse than a half-assed open-source product. Wouldn't take those "general" assumptions you make any further than I could throw them, seeing as the only successful commercial languages I can think of off the top of my head are... ColdFusion, (formerly) Visual Basic and Delphi. The only one of those which didn't build off an existing userbase for a free product (and a language with existing open implementations) was CF. The vast majority of the languages used today are based off open standards. So I guess maybe that "as a general thing" doesn't apply to programming languages at all ? My general impression remains the same as it was when it was first announced: MX2 aspires to be a language on par in general casual usage with languages like Python, combined with the fine tradition of ROM BASICs of the 80s, where you could build simple tools for yourself and even half-decent games with right out of the box.

A language which creates its own ecosystem (including jobs) thanks to its widespread usage. Commercial languages only get this if they have institutional support from other commercial entities which rely on it, commercial entities which have ecosystems of their own. Obviously this is why the only commercially successful languages tend to be ones used in enterprise or internally by large companies, and why open languages are popular with "everyone", particularly hobbyists and startups.


Danilo(Posted 2015) [#13]
> with a poorly defined feature set, lack of documentation, and bad user experience.
> The attitude tends to be "you've got the source, you fix it".

I agree that many open source projects are done like this. My experience is often much, much better with
commercial products.

But, once you know it, you can change it. That's why I think it's important to not release too early (to general public),
include at least some basic docs, IDE, etc.

Of course not everything is perfect from the start, and open source is about collaboration of many
people over time - so Mark does not need to do all himself, ideally. If some specialists jump in (write docs,
enhance the IDE, add new imports and libs), development of the project will become faster over time.
But, it still requires time. The perfect and complete package will probably not be ready within a month.

Interested people should be able to download a binary release and install/unpack it,
to see a project that ideally works out-of-the-box, without too much hassle or compiling
everything yourself before being able to test the project.

I think usability and quality is still important. The mindset should be to do a premium segment product, even if it's free open source. ;)


dawlane(Posted 2015) [#14]
@Nobuyuki: I couldn't have put that better myself. I can think of a number of closed source, paid for products that fits what most of Josh said, and people moan about it.
For open source to be a real success. It needs a well organized, well funded and dedicated team behind it.


abakobo(Posted 2015) [#15]
For me open source is the correct path for mx2. It's one of the reason i'm in patreon. I would be disgusted if it turned close source. And I will probably continue to suppport till Mark Silby retires if open source. And a that time the product should be good enough for a community to manage it.
But i hope it will get a better open-source community participation and use than Jentos for example. Wich is a phenomenon I can't understand: there's no sticky thread about it and it's not included in the distro. And the paid ide's are more visible...
I beleive mx2 can fly.. But not proprietary!


dmaz(Posted 2015) [#16]
One of the most important things for an open source or any language these day is an official module/package archive. that archive should have good cmd line utils for installation and listing and also an easily browse-able web page. I would argue that not having this has been a major issue with Mark's products gaining more support.

I would also argue the earlier the better (obviously he needs a language that can produce these module's first)