my email client is Pegasus Mail. it has been since i moved to windows, around 1996 or so (before that i used OS/2). back then it was one of the few email clients that supported multiple accounts, and managing messages on POP3 servers (i.e. downloading only headers, to allow the user which messages to download, delete, both or none).

at this point you might ask, if i translate mozilla, why don't i use mozilla mail. that's a valid question, and the answer is, mozilla mail and thunderbird still don't have on-server message management. they did half the job, and allow you to download headers and decide which message to  download. but you can't delete, so what's the point? i plan to work on that part sometime.

today, there are many email clients that do what i need from an email client, which include the two things i mentioned. i would also need a strong template system (also something mozilla mail lacks). pegasus has the strongest template system i've seen so far, with dozens of possible parameters you can put in a messge. it also has one big advantage: it can read and write in hebrew. wether this is by design or by accident, in a survay i did a few months ago, going over every email client i could find on the net, there wasn't one that could handle hebrew correctly.

but pegasus is not without problems. in fact, it has many problems. it has bugs i run into on a daily basis. it crashes. it somehow uses the keyboard language mode to determine the encoding, in complete disregard to the message encoding. it doesn't support unicode, and outside the message window, it doesn't display anything which isn't english.

so i sat down for an hour, and wrote David Harris, who develops Pegasus, this email:

subject: why Pegasus will fail

hello,
this is a long one, please bare with me.

my name is Tsahi (you will probably pronounce it wrong, but never mind). I've been using pegasus mail since about 1997, and I think that in terms of features, it is probably the most feature-complete email client in existence. even now, I still don't use all of its options. to me, it also has one indispensable capability: it is one of the few email clients that allow writing in Hebrew, wether that's intended or not.

but having used Pegasus for so long, I am also aware of it's numerous bugs. and I'm not talking about things that happen in rare situations, but things that I encounter on a daily basis. of course, I reported all of these to your tech-support about two years ago, but it's obvious to me that releasing a product with so many problems has only one meaning: the development team can't handle the size of the software.

in your web site, you write, "To me it [open source] is an unproven concept, with the obvious potential to produce low-quality applications…"
David, your application is already of low quality. it renders HTML horribly, it crashes often (i'm using WinXP, on a two months old install and SP1), it has all these bugs (7 bugs I encounter all the time, that I reported), it doesn't support unicode in an internet which is increasingly multilingual, it sometimes doesn't render JPEG images, and more. you just can't get any lower.

I still use it, though, because of its Hebrew ability combined with the "selective mail download" option which I can't live without.

now, regarding the low-quality problem: I can only guess where you got this notion from: the FUD spread by microsoft and friends, a company you describe in harsh words on your web site (words that I agree with). the fact that open source software is often written by volunteers, sometimes without formal education in programming, doesn't mean the outcome is low-quality software. I am involved in the localization of the mozilla browsers. as such, I am familiar with their development practices. the idea is, that anyone can submit patches for bugs or new features, but these patches must be reviewed by an approved software engineer before they are added to the code base, to ensure their quality. I don't know how many people currently work on Pegasus, it could be you alone. but these people can be these reviewers. this will increase the amount of code each one of them (or you) can produce – reviewing takes much less time than coding. with time, you can let other people, who have proved their coding abilities, to review code, and increase the development speed even more. this method of work so far created excellent software, such as mozilla, openoffice, linux itself, and a variety of other programs.

if you are afraid someone will "steal" your code once you open it, don't be: in the open source world, forks are relatively rare, and usually happen only when the original software development team screws really bad. or, if someone wants to take it to a new direction. but I've been told by someone that your code base contains many win16 calls and other quirks, and as such, I doubt if anyone will want to use it before it's quality is improved. if, for example, someone will want to create a linux version of Pegasus, this will also not hurt you, because you don't operate in the linux scene now, so you won't lose anything. at most, people who found pegasus on linux will also want to use it on windows too.

on your page, you solicit for donations from the public. have you gotten any? donations are very common in the open source world, but you state yourself on your site that so far you didn't receive any offers to help you maintain a closed source model. a program the size of pegasus, if open-sourced, is sure to receive contributions both in money any code.

since you already work in a business model very common in the open source world – give the software for free and charge for support, I think you are in an ideal situation to open the source, and allow the thousands of talented people on the internet to to help you make pegasus a great piece of software.

I do hope you will take what I see as the right step, before you are left with no users at all. the internet community can help you tremendously in improving pegasus, and in my view, it is an excellent way to revive pegasus mail.

sincerely,
tsahi asher.

//end message
and the answer was:

On 11 Apr 2005 at 18:06, Tsahi Asher wrote:

> this is a long one, please bare with me.

Nope, not with a subject line like that, I won't.

Cheers!

— David —

he didn't even read it! now, i'm not surprised Pegasus sux so bad.