Tip:
Highlight text to annotate it
X
Erik: How do you balance experimentation and commitment in the projects that you pursue?
Joe: I'm always experimenting. I think that's a really important part of what I do in building
products. I think it's really important to focus your experimentation and scratch those
itches. I actually think that that increases focus because I look at new ideas. When I'm
focused on a new product, a new company, I look at new ideas as, particularly ones that
don't apply to my new product. Obviously, I'm always having new ideas about the product.
I look at them as almost like cancerous because if I allow them to germinate -- I'm the type
of person that like, "That's a pretty good idea," and then I'll ruminate on it for a
little while and I'm like, "That's a really good idea." And then a little bit longer,
then I'm like, and then I start checking out of the other one. So, experimentation allows
me to kind of like satisfy my little need to explore that, and usually I'll explore
it just kind of enough to scratch the itch and then I can move on.
Erik: Could you give me a couple of examples of projects or itches?
Joe: A couple of things that I've done, I created -- for a long time, back in like 2001,
2002, I was doing a lot of consulting work and had a ton inbound email and IM from clients
and stuff and email at the time, I mean, the very best email client out there was terrible,
right? And I was like, "it would be really awesome if I could take all of my email and
kind of index it by not just subject and body, but the attachments." There's a lot of really
great data in emails. It's still the primary mechanism for sharing on the internet. There's
a lot of URLs in there, a lot of photos, a lot of videos. Obviously your real social
network is in your email at the large part. So I kind of like thought that was a cool
idea and I had always wanted to do that and ended up actually scratching an itch over
a weekend and built this thing where you could log in to Gmail and then it would index. It
was like a little internet crawler that would crawl through your email inbox and then would
basically, you could do things like show me all the photos my mom sent me last week, show
me all the PDFs that my lawyer sent me last month, show me all the Excel spreadsheets
that I've gotten from my accountant. It took me three days to kind of hack that together
and the proof of concept was awesome. I was showing it to people. I'm like, "Check this
out," and they're like, "That's really cool," because you could click on someone's name
and see all the attachments that they've sent them. You could click on a month and see all
the attachments you got in a month. You could slice and dice in all sorts of different ways.
And basically I just scratched that itch and left it stagnating on GitHub for months and
then met a guy who had just graduated with his masters in Natural Language Processing.
And of course, email is like that guy's wet dream, right? There's just gigabytes of text
and he got really excited about it and he was like, "Can I hack on it with you?" and
I was like, "Of course, it's just gathering dust on GitHub, fire away." And that company
went on. Those guys, he recruited in a friend, they went on to leave their jobs, helped them
raise 500 K and by the time this video hits the internet, they'll have closed theirs hearsay.
Erik: What's the name of that company?
Joe: Attachments.me, and it's pretty amazing now because they have taken it and completly
ran with it and they've built it now and they're like this really crazy content router. So,
you can literally say -- a lot of services send PDFs as receipts. GitHub is a good example.
So they send an invoice once a month with the receipt as a PDF attachment. So now in
Gmail, I can say, "Please route all PDF attachments from GitHub to my Dropbox GitHub receipts
folder," and it just happens automatically. You don't have to do anything. It's really
nice.