4,848 karma · joined November 17, 2009
This is why Amazon Grocery is so important - it lets Amazon build out a shipping infrastructure in a way that seems uncompetitive to UPS/FedEx
What we hope to do at IFTTT is develop a containerization of information; a common interface for moving data from point a to point b. Now, I realize we run the risk of creating the 15th competing standard [2], but we have some big ideas on the horizon of how we can make this work in a really general (and beautiful) way.
BTW, if these types of problems interest you, we're hiring :)
It would probably be better to pitch that our angle is code-heavy rather than news-heavy. Thanks for this feedback. Also, A drip of Javascript is really cool. I just subscribed.
If you're interested in keeping up with clojure news, I recently started a clojure newsletter where we cover this sort of thing: http://defnewsletter.com
We also believe in measuring everything you can. We're interacting with many APIs across many boxes. Statsd + graphite are the tools we use to understand what's happing at runtime.
Graphite has a lot of warts, but it's really powerful once you get used to it. There are plenty of pretty interfaces you can put over graphite, but nothing really matches it for ease of ad-hoc queries.
Typically I'll use graphite to view ad-hoc metrics and build reports. When I find I'm repeatedly viewing a particular graphite report then I'll "hard-code" it in gdash [1] for the rest of the team.
We use this combo to track thousands of separate metrics and we've been pretty happy with it so far.
Put the following in your ~/.inputrc
"\e[A": history-search-backward
"\e[B": history-search-forward
"\e[C": forward-char
"\e[D": backward-char
set show-all-if-ambiguous on
set completion-ignore-case on
Now if you type, say: cat <UP ARROW>
You will cycle through commands in your history that start with the prefix "cat ".Quick triggers, on the other hand, are event-driven.
So, for instance, if you had a task that was "IF I send IFTTT a Google Chat THEN post to Twitter" we'll now post your tweet within a couple of seconds (vs. before it would have been within 15 minutes)
The code is here: https://github.com/auser/beehive though I'm not sure how active it is.
Typical behavior-driven ranking features (impression vs. click-through, length of time a user is on the site, number of times visited) can be simulated with clever enough bots, relatively inexpensively.
On the other hand, flattr has is not only a record of websites viewed, but an actual amount of money must change hands and is a powerful feature for the social proof of a website.
Granted, maybe there is a sense in which this makes it easier for spammers (just throw a bunch of money at it via flattr and you'll rank higher!) and they'll undoubtedly have to deal with spam like everyone else. But all in all I think this is a clever move and no one (in my knowledge) has tied social cash so directly to ranking.
We're hiring! http://ifttt.com/jobs or email me: nate@ifttt.com
Ruby / iOS / Evented Systems
replicatorblog is going to help me by suggesting some stability and aesthetics improvements, but if anyone else cares to join in, you're more than welcome.
The slots will be cut-out so you can adjust according to your height. My goal is something that can by cut with a CNC/laser cutter and then sold for ~$20.
Does anyone have any links or resources on how to do a 'hello world' of audio-based data transfer?
Our society has grown to be incredibly complex. Imagine taking your great-great-great-great-great-grandfather and dropping him in modern times today and asking him to fill out the forms to get a drivers license, prepare his own taxes, set the clock on your dvd player, or send an email on an iPhone.
I guess my point is that the "common" man is pretty capable of feats that would be considered extremely technical from a historic vantage point.
This isn't merely about the long-form writing becoming dumbed down to button-pushing.
Once you get past the pictures of shirtless Ryan Gosling there is something really powerful: similar to how Google captured the link graph, Pinterest is capturing the interest graph [2].
[1] https://twitter.com/#!/josephflaherty/status/150961041751289...
[2] https://twitter.com/#!/josephflaherty/status/142235130872868...
With clojure, on the other hand, if you haven't done "functional" programming at a large scale you literally have to relearn "how" to program. You don't have state, you can't assign variables, you only sort-of have objects, and there are all these funny parenthesis all over the place.
So first ask yourself the question "who are you going to be programming with" and "what do you want to learn".
I've written and deployed non-trivial production systems in both and I'm fond of saying that scala drove me to clojure. I'm not trying to start a flame-war or anything [1], so let's start with the good stuff:
* you get functional constructs & concise iterators
* ruby-like mixin functionality is awesome (implicits) - this allows you to architect using composition over inheritance which is a huge win over java imo.
But...
* the syntax of scala can get quite perlish at times (my favorite being the "crying-kissy-face-operator" :_* [2]).
* the strong typing can be maddening. For instance, if you're trying to use two different JSON libraries (as in early days of Lift) you have to do all sorts of ugly type conversions. This leads to type boxing and...
* For "nothingness" tests you have: Null, null, Nil, Nothing, None, and Unit. Fun. [3]
* inline-xml (who asked for this?)
It was when slogging through a Lift-based recommendation system that I discovered clojure and I have to say, I love clojure. You can program in a lisp but still use the gobs of java libraries available. The syntax is straightforward, the performance is good/acceptable, and because it's on the jvm you can use it in Hadoop etc.
So while it does take longer for an OOP-minded person to think in more functional terms, I think you'll find that learning a lisp will make you a better OO programmer.
* [1] I have tons of smart friends who love scala. I like scala better than java, but if I had my pick, I'd pick clojure over scala. YMMV
* [2] http://www.mail-archive.com/liftweb@googlegroups.com/msg0822...
I've got a few iOS apps under my belt and I feel like I could have something to offer in this area. Feel like Skyping for a half hour or so? I'd love to help talk through your architecture. My email is in my profile.
One feature I would love to see with pinboard (or delicious) is recommended 'bookmark buddies'. That is, someone whose bookmark tastes overlap with mine (weighted by how rare the bookmark is) [1]. Recommending specific web articles is hard, but I find that following the full bookmark feed of 2 or 3 carefully selected individuals can often be more relevant than generic recommendation services (e.g. Stumble).
[1] This would have been possible on the old delicious site but delicious had severe throttling in place to prevent crawling this data. Now it appears avos has removed the list of 'co-bookmarkers' entirely from the website.
You can easily make normal people feel like their idea is worthless and you have mountains of business acumen that, obviously, they must lack. Do this enough times and what you'll find, just like you'd hoped, people stop telling you their ideas.
I'd argue that every time you do this you are shutting down opportunities to collaboratively create ideas that would probably work. Part of being "lucky" is increasing your exposure to failures [1].
When someone approaches you with an idea, you've certainly got a better feel for what will "work" more than you're average person. Consider taking the time to give feedback, help shape it, brainstorm with it. You might find that between the two of you what you create is bigger than what either one could come up with individually.
This applies in all areas of personal relationships, not just iPhone games.
[1] "To increase your chance of success, double your failure rate" - attributed to Tom Watson
Apple can be incredibly secretive and still hire with no problems because
1) they are literally creating world-changing products and people want to be involved with that
2) they have a reputation for hiring extremely bright people. If you take a job at Apple you will interact with them (/ be one of them)
But outside of the big technology names (msft, google, fb, etc.) there's a huge collection of companies where either a) the brand is known but the company isn't know for being an innovative place in the hacker community or b) the company isn't known and neither is the team.
In both of type (a) and (b) companies you hear a lot of moaning from managers right now about how they can't find any good folks to hire.
So I'll qualify my advice a little bit: If you find you're having a hard time finding "rockstar-ninja-l88t programmers" by posting to Monster.com, maybe consider that you're approaching hiring in the wrong way.