This blog is highly personal, makes no attempt at being politically correct, will occasionaly offend your sensibility, and certainly does not represent the opinions of the people I work with or for.
Communication Engines and the Timeline Problem

Part 1

I haven't been particularly good at following my RSS feeds over the past 18 months or so. My RSS reader, Vienna, has (at least had this morning) unread entries back to from 2013 (more than 1500 unread entries). Today I decided to do something about it.

Doing something about it means either:

  • mark all 1500 (and counting) entries as "read" and forget about it, or
  • just go to the view showing me all unread entries and just go to the bottom (oldest) and start reading from the bottom up (in this case I could remain 2 years behind the rest of the world for a *long* time...), or
  • every day, read all the entries of the day and also read, say, 20 past entries. Reading all entries of the day would prevent further accumulation and in total it would take me 1500/20 = 75 days to empty the backlog.

Turns out that Vienna has a "Today's articles" view so looks like the third option could work; that being said there is a problem. If I go to bed at 10pm and there are articles released between 10pm and midnight, I won't see them when I wake up the following day. What I really needed was a view of the past, say, two days. I would keep that empty while working on the archives. Unfortunately Vienna doesn't have a "Past two days" view. (I would take a "past 18 hours" view but it doesn't have that either...)

I was already contemplating writing my own RSS reader, which I did in the past (very easy todo with the existing Ruby RSS gems), when I noticed that Vienna conveniently stores its data in a sqlite file. So I wrote a script to get the list of all unread links and built the logic allowing me to keep reading the recent new ones while reading some old one at a steady rate (I update the Vienna database so that Vienna is kept up to date). Net effect, I got what I want and, cherry on the cake, the data is also fed to the program which drives my daily activities (which in particular automatically open the urls of websites I visit daily -- some don't have RSS feeds), consequently I don't need to actually use Vienna anymore (if not for keeping it running in the background so that it writes its own database for me).

If I had not been able to "hack" into Vienna, my next best solution would have been to find a way, to receive RSS entries by emails.

Part 2

The above highlights a serious problem I have with communication engines, the fact that they are silo'ed. Let me make that clear: imagine that I remember that one year ago somebody game me their bank account and I need that information right now. In my very case, because I am a control/organisation freak, I would know where to find it, but unfortunately the world has conspired to make the search as difficult as possible for people who are not freaks. The information could be in an email somewhere lost in your inbox (which email account ?), somewhere in a Facebook post, possibly in a sms on your phone, maybe in a private twitter post, or was it a WhatsApp post, or [put here the long list of silo'ed communication tools / social networks we are using everyday...].

I wish I had a single program on my computer which had its own clear format database (or APIs to make the data available to whichever program I write), and which presented me with a single timeline of all written (clear or encrypted) electronic communications that I was the sender or a recipient of, even RSS entries of feeds I subscribe to (I see myself as recipient of those) or mailing lists I happen to be a member of. In particular, every time anybody anywhere using any device and any program writes something to me, I would receive it on my unified view (and be able to reply to it from there). And speaking of communicaton, I include possible posts I could have written online (comments section of a website, message boards, etc) and the replies to me.

Sadly, I can facetime with somebody on the Internal Space Station but nobody is motivated to solve the problem of giving me a freaking timeline of UTF-8 strings.

( In fact, all this is easy to do. We don't even need to invent anything new. I could give to my phone operator, to Google+, to Facebook, Twitter, WhatsApp and all others, an email address that is dedicated to my archives, and tell them that any post sent to my accounts on their services should be copied to that email account. Unfortunately, those services/websites/apps makers are just not interested as that would disturb their business plans / revenue streams. )