Pegasus and Mercury Circling the Drain 217
Daemon Duck writes "One of the web's oldest and most respected email clients is flickering out of existence. Pegasus mail and its companion SMTP server, Mercury32, have been discontinued due to lack of funding for the ongoing development. On the website, the author David Harris states that if some funding becomes available he would consider opening the source code or continuing the development."
Opening the Source (Score:3, Informative)
if sponsors could be found to provide modest ongoing funding, I would be happy to
continue developing the programs, and would even consider opening the source.
-------------
Does it cost to open the source? It's not as simple as opening a SourceForge account and posting the source under the GPL?
Re:Never heard of it (Score:2, Informative)
It hasn't been important since Eudora was freed, and it's been totally, utterly, and completely irrelevant to all but DOS users (no idea if pmail for dos is even still around) since Thunderbird came out and made Eudora irrelevant.
pmail was highly useful back in the DOS days as it was the only free-as-in-beer client to come with a GREAT DEAL of functionality. A lot of Novell/DOS shops used it for just this reason; it even played well with Netware.
Re:Opening the Source (Score:5, Informative)
In the past, I have taken a cautious "wait-and-see" approach to the idea of Open Source. I am now willing to accept that it is a valid model, and that it is producing some genuinely excellent packages (such as FireFox, of which I am inordinately fond). Ideologically, I believe that Open Source and I are a good match, and I would like to consider going that way.
There are still some major problems with the idea of going Open Source though: the most important is "How do I survive in an Open Source environment"? While Pegasus Mail and Mercury do not require a huge amount of money to develop and support, the fact remains that they *do* require a level of funding, and I am not entirely sure how this would work within an Open Source model. I feel it is significant that the majority of Open Source initiatives are either funded externally (Mozilla), or basically not funded at all (OpenLDAP, OpenSSL): it seems to me that while Open Source is an excellent technical solution to the problem of large-scale development using widely-spread teams, the area of Open Source business modeling is one that still has not been completely resolved.
The other major issue with Pegasus Mail is that it uses a proprietary third-party product as its core editor, and I would not be able to take that product with me into an Open Source environment. The same problems do not exist with Mercury, because I have written every line of the package myself, but with Pegasus Mail, the problem is significant.
So, there you have it: I am now favourably disposed to the idea of moving towards Open Source, but have to overcome some important issues before I go down that track. I am actively considering the issues and hope I can find workable solutions (such as a large, friendly, wealthy sponsor) in the not-too-distant future.
Hopefully this update to my position will reduce the amount of hate-mail I have received in the last three years from Open-Source zealots. While I understand the passion and admire the zeal of these people, I would suggest that a positive approach is always going to work better than trying to rip out my liver and feed it to the dogs. After all, this *is* my baby - I have been working on these programs and providing them free of charge for over fifteen years now, and I don't believe it's too much to ask if I expect a little basic human courtesy.
If you have suggestions and are willing to present them to me in a positive, encouraging manner, I will be happy to receive them.
David Harris
Owner/Author, Pegasus Mail and Mercury Systems,
April 20th 2005.
Re:Never heard of it (Score:5, Informative)
So far as opening the source goes, I'd love to see it happen (actually, I'd love to see someone hire him to run it as an open-source project), but I don't know how dynamic a community could be forged around a Win32 codebase that I understand to be optimized for performance and minimum resource use over modularity, portability, and ease of future development.
Well, old dinosaurs always whither... (Score:5, Informative)
Pegasus mail was great when it started. Then a Windows version emerged, with was potent, flexible and useful, despite some quirks (you could not select anything less than a line in message text -- this gave me the habit I still follow to put URLs in e-mails on a single line without unrelated text).
But it is obviously a product that evolved by slapping-on additions haphazarldy; the configuration was nothing but unified. Related features were spread accross several configuration screens amongst several configuration options, without a grand master plan.
In the end, it was a sorry kluge that was easily replaced by other clients (Eudora, Thunderbird) who eventually evolved to Pegasus' capabilities, but without the configuration nightmare.
So it arrived at it's natural end of life. It cannot compete against nimbler and swifter clients, so it now belongs in the annals of internet paleontology as a reverable footnote, much as the Great Eastern does in steamship paleontology or The Rocket in locomotive paleontology.
R.I.P. Pegasus, you won't be forgotten, but certainly not missed.
It's time to move on. Perhaps Mr Harris could bring his expertise and experience to Thunderbird, where he would be more than welcome.
Re:long time user. (Score:5, Informative)
He would have to do a lot of work to open source it since he is using a third party editor component he is offering to take it open source if he can make a living at it. I guess he could just toss the code out to the wolves after striping out the editor but it would honestly just die at that time.
Re:nothing to do with... (Score:2, Informative)
Re:A rather hasty reaction (Score:2, Informative)
Re:well then... (Score:5, Informative)
In fact, I've just changed jobs and got a new pair of wings
Re:long time user. (Score:3, Informative)
Re:long time user. (Score:4, Informative)
> keep up with competitors. Nearly everyone has already switched over to Thunderbird
> or something else
Failed to keep up? You're out of your chair.
If you compare Pegasus and Thunderbird side-by-side, Thunderbird looks positively feature-impoverished. If people have switched from Pegasus to Thunderbird, it is because they were no longer willing to be tied down to MS Windows. Running on other operating systems is the *one* meaningful feature Thunderbird possesses that Pegasus does not.
Going the other way, there are many downsides to Thunderbird, the most significant being that its filtering is nowhere near the same ballpark with Pegasus Mail's filtering. It doesn't have flow control. It can't filter based on status flags like has-been-read, has-been-answered, or cetera. It can't filter based on time elapsed since receipt (e.g., leave unread messages in the inbox for up to ten days, then move them to another folder based on these rules...). It can't highlight a message so that it shows up a different color in the list. It can't form-reply. It can't launch an external process to handle certain messages. Et cetera, ad infinitum, ad nauseam, ad bedlam.
Besides the filtering, there are a number of other useful features missing from Thunderbird too, but that one is really the biggie. If your mailreader can't presort your mail for you, whatever else it's going to do to save you time is going to pale by comparison.
Heck, even *Gnus* (the mailreader that has every feature *including* the kitchen sink, and a learning curve to match) is missing some of Pegasus Mail's more useful features, features that Pegasus Mail has had since 1996.
Pegasus Mail didn't fail to keep up, in terms of development or features. What happened is two things:
First, and most important, it failed to be ported to the operating systems that are used by power users who crave powerful software with powerful features. A great many former Pegasus Mail users no longer use Windows. Wine didn't mature fast enough, and users were forced to find other mailreaders that would run on the OS they wanted to use. Choice of mailreader, even a really great mailreader, was not a strong enough factor to drive the choice of operating systems.
Second, it lost the end-user market when operating system makers started bundling cheesy half-baked mailreaders with the OS. But it shares that trait with most other mailreaders, except for the bundled ones and the ones that were never aiming for the end-user market in the first place (e.g., Gnus).
I understand why *development* of Pegasus Mail has stopped, in the absense of funds. But it would be nice if it could continue to be distributed for another couple of years. It's still quite a good ways ahead of the dev curve. Thunderbird will catch up with it in time, but based on its progress to date it could be another five or ten years.
Meanwhile, I'm now stuck with no decent option to recommend to people who don't want to fight a learning curve (to adopt e.g. Gnus) and aren't satisified with Yet Another Lame Outlook Clone (e.g., Thunderbird).
If bandwidth costs are the problem, he could have just given permission for others to distribute the unchanged binaries.
I suspect what's really going on is that he hopes somebody offers to buy the source. Frankly, I hope so too. I've been hoping so for a while. If someone were to manage to port it to *nix/X11, I'd be a very happy man.
Re:Is it so complicated? (Score:2, Informative)
Here's a link to his official statement about Pegasus Mail and open sourcing it:
http://www.pmail.com/sundry/pmlinux.htm [pmail.com]
I corresponded with him for a while, exploring the possibility of open sourcing it. He really has problems letting anyone else touch his code. I think it's almost a phobia, or just an attitude that "Nobody else could possibly code as well as he [thinks he] does."
Also, he seems to have little or no experience working as part of a coding team, and constantly mentioned the "communication problems" he'd have to deal with. He also pointed out that the code would need cleanup before others could make progress on it, but (to me) that begged the question of why he didn't clean things up long ago - Pegasus had a good share of bugs still unpatched in later versions when I quite using it somewhere in v3.2.
He also only wanted to "open source" it on his terms - a far cry from any existing Open Source licence. From my discussions with him, his idea of a license is: The code is his property, Anything another coder adds to it is also his property, Only he can distribute the product (including make any money from it). As I mentioned to him, that's not an attitude that will go anywhere in the open source community.
Finally, as he mentioned in an article he had on his site - he doesn't think that any quality product has emerged from any open source project. Since his only experience is with DOS, Novell, pre-OSX Macs, and Windows - I'm not surprised.
He did amaze me for years - singlehandedly developing and supporting Pegasus Mail (for DOS, Netware, Mac OS, and Windows), the Mercury mail server (for Windows and Netware), and related utility programs.
Sometime back, pre-OSX, he decided to drop the Mac version despite the fact that a large chunk of his userbase was Mac users in educational environments. He dropped it because Apple dropped supporting Netware - but he could have kept developing the product for Mac and just have his Mac users user its existing standard Internet mail capabilities. Then he dropped the DOS version, then the Netware versions. Now he's dropping the whole thing.
I think the root of the problem is that he has refused for years to bring any other coders into the work, and now is reaping the results.
FWIW, his website contents are still up there, accessible through Google.