Recruiting companies often get a bad rap because there are so many of them just spamming everyone all the time. But, as if to prove the main point, there are always exceptions, and I think I may have just found one.
While multitasking I am completing the registration process for Toptal. It actually looks great, I've heard good things from people in the industry, so I'm finally saying: why not? It seems they are all about the recruitment process, something which I have done and know first-hand how painful it can be, so I'm very curious to see how a big player does it (and, well, whether I fit into the top 3%).
My part in recruiting has always been technical. I had the task of asking unsuspecting developers to write a certain piece of software, right there in a Google doc, and evaluating their performance afterwards to come up with a (hopefully binary) recommendation. This is sometimes controversial, since you can't know for sure the candidate's abilities through a single high-pressure conversation, but it's often a good indicator, and when combined with other steps (interviews focused on soft skills, mini-projects sent as homework, etc.) it's more useful than not.
Recruitment is hard, and it's something I respect if done well. I'll follow up here if anything happens, good or bad.
20160502
20130605
Contraction counter
Yes, I'm gonna be a dad. Soon. Which means, of course, that I need a good app to count my wife's contractions when the time comes. Take a look at the stores and you will find zero good apps. How difficult can it be to build a simple, good-looking, contraction counter app?
I just started reading this book, Where Good Ideas Come From: The Natural History of Innovation, one of the central ideas, it seems, is that in order to have good ideas, one should not protect them, but rather connect (i.e. share) them. So, in that spirit, here is my idea:
Figures [A] and [B] show the app in the "Record mode". In [A] the user can tap the big circle to start timing a contraction. [B] shows the app when a contraction has started: a big minute/second counter at the center, a smaller pulsating circle, and a change in the background color, until the user taps it again. After [B], the user is presented with a popup (not shown here) in which they can select the intensity, save the record, or dismiss it.
At the top of the two first mockups you can see two graphs. [A] shows the time between contractions, and how it gradually decreases. The graphs are always updating, since "now" is always shown to the right. The labels in the x-axis mean "time ago", for example "2h" means "two hours ago". Each dot on the line represents a single recorded contraction. In [B], the graph represents how long the contractions are. Typically, this value will increase over time, rather than decrease.
The last mockup [C] shows the "Log mode". Here you can see a simple list of records, and you can edit/delete them (not shown).
So there. Simple app, simple interface. Crappy design, for now (I could certainly use some help here).
As for platform choice, I think the best idea is to build the app using one of the many cross platform frameworks, or even HTML5. The only tricky part is the graph, which would ideally let you pinch-zoom to see a variable time range.
Hopefully a basic version of this will be built very very soon, but I'd love for others to add to my idea, so it becomes a truly great app.
I just started reading this book, Where Good Ideas Come From: The Natural History of Innovation, one of the central ideas, it seems, is that in order to have good ideas, one should not protect them, but rather connect (i.e. share) them. So, in that spirit, here is my idea:
At the top of the two first mockups you can see two graphs. [A] shows the time between contractions, and how it gradually decreases. The graphs are always updating, since "now" is always shown to the right. The labels in the x-axis mean "time ago", for example "2h" means "two hours ago". Each dot on the line represents a single recorded contraction. In [B], the graph represents how long the contractions are. Typically, this value will increase over time, rather than decrease.
The last mockup [C] shows the "Log mode". Here you can see a simple list of records, and you can edit/delete them (not shown).
So there. Simple app, simple interface. Crappy design, for now (I could certainly use some help here).
As for platform choice, I think the best idea is to build the app using one of the many cross platform frameworks, or even HTML5. The only tricky part is the graph, which would ideally let you pinch-zoom to see a variable time range.
Hopefully a basic version of this will be built very very soon, but I'd love for others to add to my idea, so it becomes a truly great app.
20130228
Coding education
I think the distinction between "everyone should learn to code" and "everyone should have the opportunity to learn to code" is an important one. If you say "everyone should learn", the next thought might as well be "what's so special about coding if anyone can do it?", which leads us to dangerous terrain.
20120512
Leyes del Ecuador
Esta es mi primera aplicación de iOS. Lenguaje nuevo, ambiente de desarrollo nuevo, sistema operativo nuevo...
20120105
Basic security
How often does one have to run across basic security issues being neglected in our beloved Internet, in daily use?
Too often.
Too often.
20111030
Levels of abstraction
The other day I was .. "surfing the web" (which these days means looking at an endless twitter feed) when I stumbled across an interesting (and very long) post about a programmer's career and how to not screw it up, basically. One point Patrick made was about communication. Let me copy the paragraph and discuss it later:
20111017
A Github portfolio
In many professions, having a portfolio is a really good way of letting your work speak for yourself. A portfolio of your past work can help a potential employer decide whether they are going to hire you, and colleagues see what you're made of, and it often much more useful than a CV. Programmers can have portfolios too, in the form of public repositories, which is why I'm starting my own using Github.
20110521
God's chair
It's May 21, 9:16pm, and a big event hasn't yet occurred, but hey, have some faith, we still have a couple of hours. God, as I am right now, is probably just chilling in his Aeron chair.
20110505
USB Monitor
It's been a couple of days since I got my brand new iMo Mini-Monster Touch, which finally brings my monitor count to a grand total of three. I had tried the old-fashioned "get an extra video card" setup, but for some reason my computer didn't start at all (not even the BIOS) when I had the second card installed, even if both cards are supposedly compatible with my Intel motherboard and between themselves (screw you Nvidia).
20110328
Validation in Zend Framework model
The model part of the Zend Framework is very open. In fact, ZF doesn't help you at all with it, other than providing some classes that manage a DB connection.
Which is how it should be, really.
The model is the base for your application. It should have all the so called "business rules", and this includes, of course, validation. Now, ZF has some pretty neat validation classes that you can use. The problem is how to write the code only once for the model and for the forms (view).
Zend_Db_Table is good and all, but if you want a real-life model you need to work from scratch.Which is how it should be, really.
The model is the base for your application. It should have all the so called "business rules", and this includes, of course, validation. Now, ZF has some pretty neat validation classes that you can use. The problem is how to write the code only once for the model and for the forms (view).
20110320
Biometric login
In Ecuador we have this bank, which we will call .. Banco del Pichincha, since that is its name and I have no intentions of masking it. My dear bank is arguably (very, though they are quite confident about themselves) the most reliable and stable bank in the country, or at least one of the most reputable ones.
So a month or so ago they decide to change the login system for their website, which means people have to register again and answer a bunch of questions about themselves if they (we) want access again. So far so good, we are promised a new login, which basically consists of a username and password instead of your id number and the incredibly unsafe 4-digit PIN. The new system is called nothing less than "biometric login". Now, I assume anyone with a slightly greater-than-average curiosity will try to figure out exactly what it means, but there's no real explanation on the website other than something like "it magically knows who you are using a digital fingerprint, of your damn soul". (I'm paraphrasing)
So a month or so ago they decide to change the login system for their website, which means people have to register again and answer a bunch of questions about themselves if they (we) want access again. So far so good, we are promised a new login, which basically consists of a username and password instead of your id number and the incredibly unsafe 4-digit PIN. The new system is called nothing less than "biometric login". Now, I assume anyone with a slightly greater-than-average curiosity will try to figure out exactly what it means, but there's no real explanation on the website other than something like "it magically knows who you are using a digital fingerprint, of your damn soul". (I'm paraphrasing)
Subscribe to:
Posts (Atom)


