Re: Liberty-eiffel Digest, Vol 50, Issue 5

[Peter Lubke]

I have followed Eiffel since the early 1990's. Around 2003 when I was heading up a programming group and we were moving toward .NET as our new customer standard base set I *almost* moved us in the direction of Eiffel. It looked promising for commercial strength development *soon*.

However, around that time there were several setbacks for Eiffel (standardisation meltdown, open/not-open, pseudo-.NET and abandonment) which in the end justified the move to C# instead. Not that I liked C# better, it's just that under the restriction of .NET it was (and largely remains) the only real choice. Moving on though.

In 2007 and later in 2011 and 2015 I revisited .NET for Eiffel even going so far as to develop proof of concept compilers to produce native CIL. The issues for those attempts lie mostly with .NET support, however the later "standards" of Eiffel are also problematic. Going back to an earlier "version" of Eiffel would give it more relevance. Hence Liberty Eiffel as a potential source base - licensing, openness and cleaner language.

Today though, it's 2021 and I'm not as motivated to revolutionise the world, plus there are newer languages with moderate to good support that occupy a similar niche to Eiffel. I feel that Eiffel under the ?guidance of EiffelStudio has wandered all around the shop for many years and hasn't really contributed to any increase or improvement in Eiffel capability. Just new clothes and frequent changes in fashion.

While .NET has solved some of the issues that have plagued it from the beginning (finally after 20 years), one important one remains although it seems they are working on that too after all this time. However, it leaves other languages many many years behind C# and while libraries written in different computer languages are not supposed to harder to access, there really hasn't been any support testing or proof for that.

Now EiffelStudio aren't interested in going there because, well simply put any solution for Eiffel .NET that does work well would just plop right into the standard Microsoft tool chain FAR better than EiffelStudio which would just kill any revenue stream they are getting. So, I understand that as a commercial idea it's suicide.

So essentially, except as an academic exercise, it's pretty much a lost cause.

From an academic point of view, languages like Scala and Rust have far greater potential in terms of maturity, modernity and fitness for commercial purpose than a 35 year old concept which hasn't weather the test of time well anyway.

That doesn't mean it's all done and dusted however. It's just that a very clear set of goals and objectives would be needed to really drive any development forward. Truly it's been the lack of a clear and coherent road-map for the future that's put Eiffel where it is now (driven by a commercial vendor for profit in tool sales).

So if it sounds like 25 years of frustration at promise always just slightly out of reach, then I guess that's where I stand.

I could help, but I no longer have the inclination to lead such a venture.

Liberty Eiffel was my start point for my last exercise with Eiffel in 2018. 

I will follow your deliberations!

Peter

------------------------------

Message: 2

Date: Thu, 25 Nov 2021 11:57:59 -0700

From: Duke Normandin <dukeofpurl@gmx.com>

To: liberty-eiffel@gnu.org

Subject: Re: Noob First Post

Message-ID: <20211125115759.f1d06f7c3f95d5a32a16a638@gmx.com" rel="nofollow" target="_blank">20211125115759.f1d06f7c3f95d5a32a16a638@gmx.com>

Content-Type: text/plain; charset=US-ASCII

On Thu, 25 Nov 2021 19:50:31 +0100

Raphael Mack <ramack@raphael-mack.de> wrote:

> Actually there was some activity years ago, but without active

> users it was not really fun anymore to continue developing

> Liberty. You know how it is with free software projects without

> paid developers, if we can't keep the contributors motivated (by

> whatever means) it will eventually die. I for myself did not yet

> decide to finally leave and still have interest but there is also

> a real life...

>

> That said, I would be happy to resurrect LibertyEiffel, but I

> cannot do it alone. Even more these days, as I professionally

> quit from software developing I might be more attracted/motivated

> to work on it in my free time. So if anyone is interested I'd be

> happy to not have paid the domain fees for nothing :-)

I'm just an amateur, hobbyist code monkey, but would be willing to

help out. I could be assigned some tedious tasks to perform. We

could give that a try if it would help.

--

Duke

** Text only please. Bottom post is best. **

------------------------------

Message: 3

Date: Thu, 25 Nov 2021 14:17:38 -0500

From: Louis M <eiffel@tioui.com>

To: liberty-eiffel@gnu.org

Subject: Re: Noob First Post

Message-ID: <3f353733-aeff-1955-7c19-e6a9abb3dee4@tioui.com" rel="nofollow" target="_blank">3f353733-aeff-1955-7c19-e6a9abb3dee4@tioui.com>

Content-Type: text/plain; charset=utf-8; format=flowed

Hi,

I must admit that the timing is quite good. I have worked a lot with

EiffelStudio and have developed a lot of things for it (a lot of

libraries actually). But Eiffel Software seems to want to close the door

to the openness of EiffelStudio. We have to work with a 2-year-old

version (19.05) so that EiffelStudio want to launch without the

necessity of having an online account. So, I am a little bit mad about

that. But I love the Eiffel programming language. So if you need my help

with Liberty Eiffel, I would be glad to help. I am a pretty good

programmer, that beleive in open source software, and that has

experience in language development. So count me in.

Good day,

Louis M