the codist - programmerthink

Email Software, The One Thing No One Works On, and Everyone Hates

Published: 10/28/2012

I have several gmail email addresses, and managing them is getting to be a major pain. My main and oldest email address is barely usable anymore, as dealing with so many businesses and looking for work in prior years has left me battling a monster of daily email.

Now combine that with Apple's terrible Mail app which seems to think gmail IMAP is some kind of demon (which is likely is). I wind up with so many folders I can't really control, with many emails appearing multiple times hiding in all sorts of places. It's a big mess, and there is (almost) nothing I can do to fix it.

No one likes email, but it is still a necessary evil for lots of reasons. It's not obsolete and unlikely to ever be replaced by twitter or texting or smoke signals. Yet there are no easy ways to wrestle it into submission, and as far as I can tell, no one is spending any energy making it work better.

If you are on Windows you use Outlook, which sucks as much or even more than Mail. Neither is all that good at managing email from many email accounts, much less across various email providers and especially gmail. Another option of course is to use web based email, like gmail or hotmail, but they still suck unless you have only one email address. Maybe most people only have that one email address and I am just odd in that I have 5 and everyone is happy but me.

Today I tried to consolidate my five gmail addresses into a single one, with the others combined using Google's Mail Fetcher but it seems unable to understand its own mail. No matter what I tried it would not fetch another gmail account via pop. Apparently many people have hit the same problem but like so many things at Google, there is no support and no one cares there. People grasp at random voodoo on various forums hoping to find the magic incantation. So far, no luck for me.

So being the programmer that I am I started to see if I could build a basic email application. Since I only use gmail for my personal accounts I figured I might be able to build something simple on OSX for my own use. There is a Cocoa framework called MailCore which looks promising if it works.

I almost gave in and went to use Sparrow about the time it was eaten by the giant Google Black Hole. Of course the Sparrow people were apparently almost out of business as no one actually wants to pay for email applications no matter how terrible the free ones are. This likely will dampen anyone's enthusiasm to build a new email client. Even Mozilla has apparently canned their free email client, Thunderbird.

So sucky email programs are the rule of the day, and the economics of making new ones means no one will bother to build something better, much less gain any VC money or Angel money or Lemonade Stand money to try.

Of course my own time is my own time so maybe I can make something that at least I like. For me, I want to easily toss emails into piles, one for people I know, one for people I dealt with in the past, and one for random crap. It would be handy if it could figure out which pile it goes in at the start. It should never allow a message to appear more than once. Mails should be further grouped by domain, or subject (if I can figure out what it's about), or likelihood of my wanting to read it (who knows what criteria that might be). I figure once I can collect all my emails from my 5 gmail accounts, I can play with classifying them. I want something that I can easily try all sorts of different schemes too without locking me into anything. The mails should go into a database of some kind, not the idiot individual files that Apple Mail uses.

I expect it's not all that easy, but it might be fun to torture myself with whatever evil complexities I don't know about yet.

I am sure no matter what it becomes it will never be a commercial product so it only has to make one person happy, me. If it sucks I can always just use Mail.

Or I can cancel all my email accounts and move to Antigua and live on the beach. Decisions, decisions...

Tag:

submit to reddit submit to hackernews