Disqus - Latest Comments for jfarmer

Re: Can Silicon Valley boot camps get you a $120K job?

Jesse Farmer — Fri, 11 Oct 2013 15:34:22 -0000

As I said, how could I possibly argue with your statement that "an adult not wearing shoes in a room with other adults...is not the professional type of individual I would actively seek?" Your hiring standards are yours, whether I think they're sensible or not. I have no interest in convincing you otherwise, so if that is your point we can leave it at that.

However, I suspect you're being coy. I took your original point was something more like, "That person is behaving unprofessionally and it will hurt his chances at finding a job. What kind of school—especially one designed to help people find jobs—would not address that? I now suspect that there are other breaches of professionalism in the environment and question how fit the graduates of this program really are. This reaffirms my more general belief that such a program is fundamentally unserious." Now, I admit that might not have been your point. Since you so cleverly cloaked your point in sarcasm rather than stating it plainly, I hope you'll excuse me if I misunderstood it.

If that was your point or at least the spirit of your point, well, my response is two-fold. First, if you're right then programs like this will suffer in the marketplace, either because their graduates won't be able to find jobs or because other programs will come along that are "more professional" and edge out the less professional programs. If there's where it ended you'd still be left with the ethical issue of graduates who paid tuition but didn't get the level of professional education the marketplace demanded. Speaking for myself, we took feedback from employers very seriously and if we received feedback that our students were not behaving professionally we'd address it immediately, both with the individual students and the class at large. Keep in mind that none of our students to on to work as attorneys or in multinational oil companies, so those are not the companies whose feedback we care about. In fact, we've had a few dozen attorneys (who worked at firms like Orrick and WSGR) come to us!

Second, the culture and standards of your world are vastly different than that of the technology world. For example, I'm commenting here using my real name with a profile that links to my real information and you're commenting pseudonymously. Although I don't know if this is the case, it wouldn't surprise me if as an attorney you're doing it as a matter of professionalism. I think that's a sensible thing to do as an attorney and might consider an attorney who was floating around the web leaving snarky comments with his real identity to be behaving unprofessionally. If it were a partner at a firm I was working with, for example, I might think to myself, "Is this guy going to spend time handling my professional concerns or commenting around the web and stirring up trouble? I'm hiring an attorney to remove risk from my world, not to add it. I don't care if he leaves comments, I just don't want them connected to me." However, in the world of technology it doesn't make any sense at all. Nobody in my universe would ever in a million years consider what I'm doing here unprofessional nor would I ever be worried about it complicating future professional interactions.

I would put a student not wearing shoes in a classroom environment in this same category. Note: I mean specifically this sort of classroom, not any classroom in general.

Anyhow, this was a fun intellectual exercise, but I'm out! If that wasn't your point, well, we'll have to live like most people do: in a state of mutual incomprehension. I'll leave you with the last reply if you're so inclined.

Re: Can Silicon Valley boot camps get you a $120K job?

Jesse Farmer — Fri, 11 Oct 2013 14:08:57 -0000

I can't speak for other programs, but DBC has always selected students based on their interest in programming. If a student doesn't appear to be sincerely motivated to learn how to program, they'll burn out in the first week—it's just too intense.

Re: Can Silicon Valley boot camps get you a $120K job?

Jesse Farmer — Fri, 11 Oct 2013 13:56:33 -0000

I'm the person standing in the upper-right corner of the photo and I said "that's Gavin Morgan" as if he were a real person with a real name. Anything more than that is rubbish you read into my comment to bolster the point you wanted to make. Compare that to, say, someone making pointlessly snarky comments on a Fortune article using the name and photo of a character from The Office.

I'm sure your comment that you wouldn't want to give him a $120k/year job is accurate. I don't know how I could possibly challenge that. After all, who knows better than you what your hiring standards are?

I might question whether basing a hire/no-hire decision on a photo of them not wearing shoes in a classroom environment is an effective filter, but if it works for you I'm sure it's grounded in your "expertise and experience." As for me, I will continue to make hiring decisions without considering candidates' footwear choices outside work, as apparently Pivotal Labs and the other companies that gave Gavin offers did.

I appreciate you taking your valuable time to engage in such a thoughtful discussion with me.

Re: Can Silicon Valley boot camps get you a $120K job?

Jesse Farmer — Thu, 10 Oct 2013 15:19:24 -0000

Normally I wouldn't comment, but that's Gavin Morgan. After Dev Bootcamp he got a job as a software developer at Pivotal Labs.

I appreciate you going for the easy upvote with gusto, though! I'm sure your additional comments will be equally thoughtful.

Re: <em>Breaking Bad</em>: Was That a Happy Ending?

Jesse Farmer — Mon, 30 Sep 2013 14:44:51 -0000

Don't you remember the episode where they put a powerful electromagnet in the van to destroy evidence? I thought of it as a callback to the more "zany" episodes.

Re: The San Francisco Rent Explosion

Jesse Farmer — Thu, 18 Jul 2013 17:00:33 -0000

I'm in a similar situation in that I have a 2BR in Hayes Valley which I pay $2,000/month for. But that low price is subsidized by higher prices everywhere else in the city. This is the perverse thing about rent control: it incentivizes me to help myself at the expense of everyone else.

Re: Too DRY - The Grep Test - Zero Wind :: Jamie Wong

Jesse Farmer — Fri, 12 Jul 2013 23:51:29 -0000

No problem! I think one of the keys to a good explanation is removing every unnecessary dimension. It reduces the "surface area of misunderstanding," so to speak, and makes the distinctions you're trying to draw more clear.

One way I explain this technique to engineers is by analogy to a "minimal test case." When you find a bug and want to write a regression test, your goal is to identify the smallest bit of code from the broken codebase that exhibits the bug and then write a test around that. When your test case is too broad it's impossible to know whether you've really isolated the bug or not.

Here, you want the "minimal explanatory example," that example which encapsulates the entirety of the point you're trying to make but no more. This is often impossible, but the more of the world your example encapsulates the more likely a reader is to be confused about what the example is meant to illustrate, even if you point it out to them, surround the example with text telling them what to pay attention to and what not to pay attention to, etc.

Like you said, between examples #2 and #3, example #3 just seems too clever by half. What are you trying to do, save two lines of code? For what? It's harder to understand and, as you say in this article, you can't use standard Unix tools to search for method names.

I think it makes the case for the Grep Test almost self-evident.

Re: Too DRY - The Grep Test - Zero Wind :: Jamie Wong

Jesse Farmer — Fri, 12 Jul 2013 23:25:48 -0000

Ok, well, it's good that you saw the issue! :D

Now we're just talking about what makes a good example. Having that extra dimension in there makes it harder for readers to see your point, not because they get caught up on it, but because that's the dimension crying out for refactoring.

Doing it idiomatically isn't much more work and it saves you the potential trouble of being misunderstood, cf. Wiio's Laws.

Here are some slightly more idiomatic examples in Ruby, two of which don't pass the Grep Test: https://gist.github.com/jfa...

Notice I'm not using method_missing and as a result reclaim a fair amount of discoverability.

Re: Too DRY - The Grep Test - Zero Wind :: Jamie Wong

Jesse Farmer — Fri, 12 Jul 2013 22:07:41 -0000

Some more thoughts. One of the major aims of refactoring is to make code more modular. DRY is not a process so much as it is a heuristic for identifying code that might be worth refactoring. By "more modular" I mean "removing unnecessary coupling."

Taking the Ray class, again, there are two kinds of coupling. One is between the names, e.g., ray.position_is_zero?, ray.direction_is_zero?, etc. and the other is between the actual algorithms at work, e.g., "is this a zero vector?", "invert this vector!", etc.

The latter is a much more dangerous kind of coupling because it means that any time the work being done in one location changes you have to remember to reflect those changes in the second location. Moreover, anyone else reading the code has this additional, pointless "fact" they have to remember, viz., this work is being done in two places.

Name coupling happens all the time and is the least interesting kind of coupling. Any time you make an API call, for example, there's some amount of name coupling. If the name of the API endpoint changed your code would break unless you changed it, too.

Unlike algorithmic coupling, which virtually always calls for applying DRY, name coupling calls for DRY on a more case-by-case basis. For example, if two short methods with similar names are defined in the same location in immediate succession, there's little reason to remove that duplication.

So, I love the idea of the Grep Test, but the funny thing is that you solved the less-serious name coupling problem but left the more-serious algorithmic coupling problem alone.

Re: Too DRY - The Grep Test - Zero Wind :: Jamie Wong

Jesse Farmer — Fri, 12 Jul 2013 21:41:58 -0000

Hmm. I don't think the lesson here is about to-DRY-or-not-to-DRY your code. Instead, it's about using the appropriate abstractions.

Using the Ray example, both position and direction are vectors, not points. Make them vectors! Then you'd be able to say things like


ray.position.zero?
ray.position.inverse # We'd usually say "the inverse of" not "the negative of"

Furthermore, if you wanted to define the same methods, well...


class Ray
  def position_is_zero?
    position.zero?
  end
  
  def direction_is_zero?
    direction.zero?
  end
end

There's less repetition, now, because you're only repeating names, not logic. This means the code will only break when the names change, i.e., zero? becomes is_zero? or something.

In a world where all of your logic is also duplicated in the Ray class, the code would break when either the names or the logic changed.

Re: Ed-Tech [Expletive Deleted] #Edinnovation

Jesse Farmer — Sat, 04 May 2013 17:34:59 -0000

Well, if it helps, I'm one of the co-founders of Dev Bootcamp. I had an interview of CNBC earlier this week and when they asked me about our "new and innovative model" I said it wasn't anything different than Dewey was doing at the Lab School 120 years ago. If someone is familiar with M

In fact, our teacher and student handbooks have an educator's quote at the start of every chapter. Here's one I particularly like that expressed many of our values at DBC:

"There is an image of a school growing up in my mind all the time; a school where some actual and literal constructive activity shall be the centre and source of the whole thing, and from which the work should always be growing out in two directions — one the social bearings of that constructive industry, the other the contact with nature which supplies it with its materials.

— John Dewey to Alice Chipman Dewey and children, November 1st, 1894, John Dewey Papers"

Re: Orwell and the Not Unblack Dog

Jesse Farmer — Sat, 20 Apr 2013 22:41:01 -0000

I think you're missing the forest for the trees. And yes: that cliché was deliberate. :)

Orwell's core argument isn't against particular words or turns of phrase, rather it's against unclear, muddy writing, especially when used by politicians to hide what they honestly believe. By default "not unintelligent" is less clear than the alternatives. Like you said, it expresses a whole continuum of possible meanings.

If an author doesn't have a particular reason to use that phrase then they're just being thoughtless. Similarly, a politician could use it in a way to mean many things to many people and therefore avoid responsibility for his actual beliefs.

I'm thinking of a politician saying, e.g., "Unlike Barack Obama, I share America's values." That politician hasn't said a single thing about what they believe. We only really know that they disagree with Obama about something. And how can a country have "values," anyhow?

That phrase means something totally different to a crazy racist and a middle-of-the-road Republican, but each person still leaves the speech believing the politician agrees with them.

Conversely, I have a hard time imagining Orwell objecting *per se* to something like this:

John: "Joe, is Tim intelligent?"
Joe: "Well, he's not unintelligent." (emphasis on *un*)

In this case, the phrase deliberately parallels the previous sentence and makes Joe's beliefs about Tim and the nature of their relationship more vivid than a matter-of-fact "Tim is of average intelligence."

Re: The Universal Notebook: William James can take Ayn Rand any day

Jesse Farmer — Sun, 17 Feb 2013 16:35:37 -0000

"Truth then is relative. There is no objective reality, no absolute truth. We all know this in our bones, but we have a hard time accepting it in our minds."

Oh heavens, no!

"Scientists, of course, would like us to believe that there are immutable and discoverable laws of nature, that only verifiable, replicable truths are valid, but science only describes the “how” of life, not the “why.”"

Double, double no! Quadruple no! This is the exact opposite of Pragmatism. Its essence is the scientific method applied to epistemology!

It's worth pointing out that many of the core thinkers involved in developing Pragmatism were also involved in developing modern psychology, mathematics, educational theory, and jurisprudence -- at least in the United States. CS Peirce, Oliver Wendell Holmes, Jr., John Dewey, etc.

Pragmatism is post-Darwinian empiricism applied to epistemology. It says at any given moment, given all the dimensions and facts of this world, there is some true state of affairs but we can never know it in its entirety with absolute, iron-clad certainty, accuracy, and precision. Moreover, we don't know which aspects are unchanging and which aspects are fixed, e.g., is the truth of some proposition X actually a function of time, but changes so slowly with respect to time that we couldn't possibly know that?

The absolutist says "Well, either X varies with time or it doesn't and we must figure it out!" The pragmatist says, "Let's assume for a second X varies with time. What does that imply? Do any of those implications contradict anything else we believe or observe? Does assuming X varies with time explain things that were previously unexplainable?"

That's not the same as saying we can't know anything with iron-clad certainty, nor is it the same as saying everything is relative. Pragmatism is a statement about our ability to comprehend and the difference between the Truth and what people say when they use the word "truth."

Put more plainly, it goes like this. Forget the idea that we're searching for the Truth, whatever that means. Instead imagine we're building a model of reality. The constellation of beliefs that help us thrive are the beliefs that more accurately model reality. That is, the more our model agrees with reality the more we get along with reality.

If there's a core assumption to pragmatism (and science), it's this. I'll leave it to the metaphysicians to debate whether that's a valid axiom, but nevertheless that's the axiom.

This is similar to Darwinism: the species which adapt most ably to their environment are the ones which thrive. In some sense that "environment" is encoded in the the species' behavior. You can have behavior which is over fit, e.g., a sea creature that can only survive off volcanic vents miles below the surface of the ocean. You can have behavior which encompasses ever-expanding environments, e.g., humans who can self-reflect, learn, build tools, communicate, etc.

That's not to say the over-fit species is somehow wrong or bad or inferior. Compared to humans, who can't survive on the ocean floor (yet), they something going for them -- even if they can't leave a 10-meter radius of the spot on the ocean floor where they're born.

The pragmatist approach to epistemology is a bit like having thousands of species at our disposal for analysis and reverse-engineering the environments to which they were adapted. As if there's some Platonic land of Truth, but we only ever see -- and see inaccurately -- tiny slivers of that Truth. We have to take those slivers and from them be be precise about what we can say, what we can't say, and how certain we can be about the goings-on in Truth Land.

It's also strongly related to the development of statistics, which was happening at the same time. Statistics is the science of precise measurement of imprecision. We can state precisely the extent to which our measurements are accurate. We can state precisely the extent to which we're justified in believing something, e.g., there's a 1% chance we're wrong in believing X given what we know. New information might change that prior probability.

X is either true or not, we just can't say. We also don't know, for example, whether X varies according to a more global variable that appear unchanging from our perspective due to our more local frame of reference. This this similar to how even though the Earth is curved, when we're walking around on it we're justified in believing and acting as if its flat. Or how even though Special Relativity tells us that time dilation, length contraction, etc. occur, when we're moving at slower velocities relative to the speed of light we can act "as if" it isn't happening.

It's not that it isn't happening, it's just that it's happening below the bounds of our perception and below the bounds of our need to incorporate that into our thought in order to achieve our aims.

This is the essence of Pragmatism. Truth isn't relative. The Pragmatist might say that talking about "Truth" is a category error. Instead we should talk about our models of reality and how we can validate or invalidate the various aspects of those models.

Sure sounds like science to me!

WRT to Objectivism, for example, it's hard to swallow for the Pragmatism not just because of its absolutism but because there's so much of reality it simply doesn't account for.

Watch an Objectivist talk about the relationship between a parent and a child, for example, or a child and their elderly and infirm parent. The explanations for observed human behavior read like the Ptolemaic description of planetary movements. Epicycles upon epicycles upon epicycles.

http://en.wikipedia.org/wik...

This doesn't mean its wrong, but it should be a sign that its incomplete. It's also a backwards process, scientifically speaking. We have a model. We observe data that confounds it. We augment our existing model to account for this new data and only this new data.

That's not science, it's religion. I'd ask any Objectivist, "What data would convince you that Objectivism isn't true?"

They'd probably tell me I'm thinking about it wrong, that Objectivism is necessarily true and shouldn't be subject to trial-by-data. I'll leave them to that; to me that's the death of thought not the discovery of Truth.

Re: Startup dudes: Cut the sexist crap

Jesse Farmer — Thu, 09 Feb 2012 18:17:02 -0000

This is a comment I posted on Facebook

I see this shit all the time, and most men don't even realize they're doing it. Lots of startups quickly converge to a culture that's little better than a frat house.

If you think this is much ado about nothing, consider this: the subtle sexist parts of your company's culture will turn off any qualified woman from working there.

And we all know how *EASY* it is to hire in Silicon Valley these days.

I'd also like to add, one of the most positive experiences in my startup life was working for Brian Sugar and Lisa Sugar at Sugar, Inc.. This was my first job at a proper startup and as a male I was in the minority.

I joined as #10 or so, and the first non-founding engineer. One third of the founding team were female. The second engineer we hired was Lydia Wagner.

From #10 to #30 or so I think I was one of only two or three male hires.

The stereotypical Silicon Valley startup is founded by N engineers, probably male, who graduated from college in the last 5 years.

Put those people in a tiny room for 10-14 hours a day, 6 days a week, for a year or two and it's kind of inevitable.

But it's also easy to stop if you're self-aware about it, and the founders make it a priority.

Re: Why Facebook Connect shouldn’t be your only sign in option

Jesse Farmer — Fri, 03 Feb 2012 13:39:53 -0000

Bijan, I agree and have seen the same thing.

However, Facebook is changing the auth dialog style soon(ish). People can opt-in to the new style now, but it will be mandatory eventually.

The upside of the new style is that it's much less threatening. The downside is that if you ask for any "aggressive" permissions, e.g., post to feed, offline access, they're split into a SECOND dialog (yes, users will be prompted twice) that is actually more threatening.

So the net will be that people who don't ask for advanced permissions will see a conversion rate boost, and the people who do will see a conversion rate hit -- most likely a big one. I'm guessing this is to make devs think twice about what permissions they're asking for, and only ask for the ones they really need.

Here are screenshots of what it looks like:
http://cl.ly/0U0u1f233d2c2E... (first screen)
http://cl.ly/0L0t343j1q0s0f... (second screen, only displayed if you ask for advanced permissions)

You can read more, here: https://developers.facebook...

Re: Michael van Rooijen (meskyanichi) - More concurrency on a single Heroku dyno with the new Celadon Cedar stack

Jesse Farmer — Mon, 17 Oct 2011 23:27:00 -0000

I just wanted to add that if you're using Unicorn + New Relic you'll have to create a new initializer that contains the following line:

NewRelic::Agent.after_fork(:force_reconnect => true) if defined? Unicorn

Re: http://damurillo.tumblr.com/post/9214057988

Jesse Farmer — Sun, 21 Aug 2011 21:21:37 -0000

If anything, I've found the opposite. SF Tech people hate FB Connect, while the average person doesn't mind.

We have an FB Connect only landing page right now converting at 40% or so from landed to connected. The only people complaining are techno-paranoid engineers from silicon valley.

Re: Why I'd never work for Google, Twitter, or Facebook - naildrivin5.com

Jesse Farmer — Mon, 01 Aug 2011 15:47:18 -0000

Sorry, how does Mozilla support itself again?

Oh, right, by sending Google search traffic.

Re: Let’s mug a startup founder!

Jesse Farmer — Wed, 11 May 2011 03:11:08 -0000

The difference being YC never expected the money to be paid back -- it wasn't a loan.

Re: Click Hacking for Fun and Profit

Jesse Farmer — Fri, 22 Apr 2011 21:36:43 -0000

I sense some judgement! :P

Re: Click Hacking for Fun and Profit

Jesse Farmer — Fri, 22 Apr 2011 21:08:41 -0000

Both great examples. Asking people to prove themselves is a good way to make them respond.

Another friend of mine mentioned asking people questions. People just can't help but answer!

Re: Click Hacking for Fun and Profit

Jesse Farmer — Fri, 22 Apr 2011 21:07:46 -0000

Yeah, sometimes the link is a headline, which is what we'd normally call link baiting.

But in other cases (e.g., match.com) it's about getting people to enter their personal information and submit a form. Or for Facepile, to like the website.

Re: Dear Foursquare, Gowalla: Please Let’s Stop Pretending This Is Fun

Jesse Farmer — Sat, 13 Nov 2010 13:24:41 -0000

Dave McClure wrote this article in April: http://500hats.typepad.com/...

Re: http://blog.appstorehq.com/post/1212703421

Jesse Farmer — Thu, 30 Sep 2010 23:54:20 -0000

AR requires an auto-incrementing integer ID as a primary key.

This seems daft, but for InnoDB at least it's the best practice because the records are stored on disk in primary key order. So if you're using a composite key or non-integer PK you'll cause the disk to seek a lot in a write-heavy environment, unless you're using an SSD.

That said, there's nothing preventing anyone from creating a UNIQUE INDEX on what could be considered the composite key. That's just one more line in the migration file.

Re: http://blog.appstorehq.com/post/1212703421

Jesse Farmer — Thu, 30 Sep 2010 13:31:45 -0000

Looping through a list and issuing one query per iteration is inefficient whether it's done through an ORM or raw SQL. Instead the query should be batched.

I don't know how AR works, but in Sequel you could do something like: AppList.multi_insert apps.map(:app_id).zip([list_id]*apps.size)