constant beta software

Developing Open Source Software Libraries

Sun, 15 Sep 2019 00:00:00 +0000

Note: This is an old post that still seems relevant. It’s also a better fit for this blog, so I’m reposting it here.

As I started reinventing the wheel in Clojure (writing a few libraries for backend web development), I actively became aware that I should think about the philosophy behind the libraries. By that, I mean the general direction and goals of the development process, something that should guide and be the goal of all kinds of software libraries ¹.

I don’t think enough is written about the subject of creating software libraries, even though it might be the most popular artifact on github. It is simply expected that the thoughts are present when the code is written and that the result follows.

So here am I, trying to change that ².

The Characteristics of a Good Software Library

A software library is a reusable piece of software that someone is going to import into a software project. We use libraries because it saves us the trouble of writing code that can be abstracted by a third party and, thus, be reused.

I will be describing sets of characteristics that libraries should have to be considered “good” ³.

User Developers and Programming Developers

I begin by making a distinction between the two extremes of developers. It is by these points of view that we shall focus.

User Developers are programmers who are simply incorporating the library into a project, with little to no interest in the internals. They probably make up the largest slice of the userbase;
Programming Developers are programmers who are interested in knowing how things were made. This can be, among other reasons, due to simple curiosity but also idea gathering/refinement or interest in contributing back.

Programming Developers should be viewed as (mainly) a superset of User Developers.

From a User Developer’s POV

When searching for a library, users will be interested in the one that solves the problem and delivers the most bang for the buck. Remember that the goal is to have the user not write the code for the functionality that the library is supposed to provide.

As such, the four characteristics that I deem as the most meaningful are:

It solves the problem. This might be too obvious, but it is to serve as a reminder that the core problem you’re trying to solve is the most important thing. But it is also a reminder for not trying to do too much. While doing too much might in a library be more about style than substance, so far the UNIX way has prevailed ⁴.

It is efficient, whatever that means in the context of the problem. Again, this is obvious and it is expected by the users. It also follows from the previous point that we might hit a crossroad when building a library, where we have a chance to go deep or go wide. At a certain level, going deep starts getting harder and harder (micro-optimizations). Going too wide though, might become a maintenance hell and lead to complete refactorings or starting from scratch.
It is not hard to use. How hard is it to use your library? Does it provide a consistent interface and no unexpected gotchas? How many functions or classes must the user be aware of? Sure, documentation helps a lot, but it would be great if most things could be avoided altogether (YMMV).
It is not hard to reason about. How hard is it to understand what your library is doing? Is it clear what problems you’re targeting? Are you leaking implementation details to the user? How many new concepts do you introduce to a first time user? This happens mostly with libraries that abstract some underlying technology, e.g., a library for AWS or for a database. Try simplifying what’s under you but also the interface that you provide to the user.

Remember, most people using your library will fall into this category. Attack these points and you’ll make a lot of people happy.

From a Programming Developer’s POV

A programming developer will always dive into the code. They are interested in knowing how things work. So how can we help them with a smooth dive?

In practice, it’s all about following good programming practices. A software library is just like any other medium/large software artifact. One should aim for a codebase that is easy to read and understand and easy to change. Apply modularity, low coupling, high cohesion, comment your code, write documentation - the whole shebang.

While commenting your code helps, when trying to understand a new codebase, it helps a lot to have architecture documentation, i.e., something explaining how the pieces are tied together. This is something most open source software projects fail at, perhaps understandably, due to the boring nature of writing documentation.

So the best way to help programming developers is to write good code and documentation. Worried? Don’t be.

Nature vs Nurture

I don’t think that it should be expected for a software library to be “good” at v0.1 or < ~100 commits ⁵. A software library might not be born good, but it can surely grow into a good library. And that’s what we should strive for.

Originally written on the 1st of June of 2014.

While I was unconsciously aware of the requirements and goals, this passive kind of thinking is fuzzy and can get dilluted along the way. ↩
Even though there are way more qualified people to do that. ↩
This doesn’t mean that libraries that fail to satisfy these characteristics are not useful. ↩
I am actually curious and interested in seeing at cases that are opposite of the UNIX way. ↩
Why 100? Sounds like a reasonable number. YMMV. ↩

Moonshots

Sun, 15 Sep 2019 00:00:00 +0000

Note: This is an old post that still seems relevant. It’s also a better fit for this blog, so I’m reposting it here.

This was going to be a rant about how the new buzzword was already being misused around blogs for ideas that won’t advance humanity in any significant way, but I’d rather not be so bitter.

You can tell that this is a word that’s going to stick. A new buzzword for the years to come.

So I’m going to ~~pollute~~ contribute with some thoughts that are nothing new, but hey, it’s from my perspective.

God-like

The human being should aim for «being» omniscient, omnipresent and omnipotent. Yes, Gods. Add omnibenevolent to the mix if you feel like it. And well, in order to achieve all this, it would also be cool to break free from nature’s constraints. You know, stuff like eating, defecating, catching a cold or growing old.

But why should mankind evolve in such a direction? Ignore the benefits that such a state would bring to us. Looking back at history, being a God has always been the great goals of some of the most ambitious men in time. I guess it’s just the natural human curiosity and desire for power.

In concrete

Omnipotent

So instead of delivering things with drones, how about having what you want instantaneously in front of you? You pick the object of desire and have it materialized in front of you.

How would such a thing work? I don’t know. Maybe you go to the futuristic equivalent of Amazon, hit select, the necessary resources are sent over the wire/air/whatever to your location’s materializer (an ultra advanced 3D printer from Star Trek?) and in a blink of an eye, it’s there. Sure, we have to solve a billion sub-problems until something like this is possible, but that’s your actual moonshot.

Human++: let’s start enhancing our bodies for real. Ever since I read X-Men, I’ve wanted to possess telekinetic powers. That cup is too far away? Just do a gesture with my hand and have the cup come to me. Or even only think about it. Where are my keys? Keys: show yourself! You get the drill. What about flying? Exoskeletons were so popular in 20th century comics (Iron Man being the best (reenacted) example), why are we not there yet? Aim for things that enhance our bodies. Transhumanism baby!

Omniscient

Instead of MOOCs, how about instant knowledge in your brain, Matrix style? Wanna learn how to code in clojure? Just click this button and have something introduced into your body (a pill, hook up a cable into your brain, whatever). Yes, we have to solve some millenial problems such as defining what knowledge is or how to be materialized into something physical that our body can understand. Or how to make it fundamentally copyable to everyone (not just words, but an actual chemical formula that when introduced into your body would make you just know).

Next on the list is a prediction/intention machine! I’ve seen some writings around this. Basically guessing the future! Yes, just like Minority Report or even better, like Fringe, where there was this guy who could calculate the probabilities of events just like the Observers! That’s the real power of data, to be able to know the future. And with enough data… you just might be able to do so. (I wonder if we already have someone in the NSA working on this.)

Omnipresent

Instead of developing autonomous cars or building your own airport to take you somewhere, how about (yes, you guessed it) teleportation? But for true omnipresence, well, I guess the NSA is working on that, right PRISM? Imagine the day where you can view the feed from any Google Glass in the world!

A little touch of humanity

Even with a lot of God powers, Man will not be able to run away from some of its most natural desires (unless we drastically change our DNA that we no longer resemble Man - something I’m sure will happen).

So how about affordable, mobile and easy to transport homes? Not caravans - make it really easy to move around and transport your belongings. Not tents - we want a bit more of comfort. Think turtle shells. Or the TARDIS. We could carry a backpack - or something smaller that could fit in our pockets like Dragon Ball’s capsules. Sure, if we could print things on demand or teleport ourselves, there shouldn’t be a need for “ownership” or “moving around”, but still, human beings have a tendency to attachment and sometimes it’s more about the trip than the destination.

Imagine the problems this could solve! People could save major money and have a level of freedom that was never seen before. Wanna live a few months or years in Thailand? Go ahead. Or maybe in Tibet? Easy peasy.

But while we’re stuck driving our cars, please please please make them smarter. You can begin with a smart display. You know, use that big transparent piece of glass in front of us to do something more than protecting. Instead of having me look at my smartphone or my GPS that’s positioned in a not really good place, how about making the information appear directly in front of me, in the screen that I’m already looking at? Bah, beaten by Business Insider who just wrote a post about this!

Then there’s the rest

Control the weather! Ye, I know, this has been under progress for years.

Stop waste/trash. Clearly we must improve our recycling methods.

Immortality!

etc, etc, etc.

In the end, it doesn’t even matter

Just wake me up when we get there.

PS: I don’t really like the pop-tone this post has. Feels like a waste of time and letters. Ah, fuck it. I might rehash it in the future.

Originally written on the 19th of May of 2013.

(Not) The Year of Email

Sun, 20 Aug 2017 00:00:00 +0000

Note: Back in 2012 I was working in the email space, trying to improve the modern-day-worker’s life (this was before pg’s Frighteningly Ambitious Startup Ideas came out and everyone started talking about email, mind you). Unfortunately we did not succeed (otherwise this blog would be a lot more popular).

Then in mid-2013 I decided to pen down the following thoughts about the space. Again, I failed at predicting the future (shocking), and the change and innovation in the space pretty much fizzled out (it would be an interesting discussion to understand why that happened).

I still believe that the field is ripe for massive change (and I guess I’m not the only one), so I’m reposting my original post here, for eternity.

“email, to me, is the most important app I use”

Someone said this in 1997, but it could’ve well been in 2013. Email has become a central element in our digital lives. Despite all the reports of email’s demise, we’ll see how startups and old dogs are pushing the envelope.

It’s getting hot in here

Email startups became cool around 2007, when Xobni kicked off a mini-revolution. Its Outlook extension quickly garnered a large user base which attracted Bill Gates and Microsoft’s attention and a rumored (rejected) acquisition offer.

Since then, several email startups have popped up and got the tech world’s headlines with their products and/or acquisitions. Yahoo bought xoopit (a social networking service based around email) in 2009, reMail (email search for iPhone) got picked up by Google in early 2010 while Rapportive (rich contact profiles inside Gmail) was acquired by Linkedin just last year. And not long ago, PowerInbox and ActivePath decided to join forces to enrich your email content.

And you know things are heating when you start having conferences around them. Inbox Love had its 2nd edition last year featuring a big number of startups side by side with the established players. Heck, even VCs are jumping in.

Innovation in the email world

There’s a multitude of startups trying to improve our email experience by either extending existing email clients or completely replacing them.

Email++

Startups that are trying to extend the email experience are doing so by standing on the shoulders of giants like Gmail and Microsoft Outlook. So what types of products are they building?

Are you looking for something that might make you more productive? Baydin’s Boomerang tools allow you to, among other things, schedule an email to be sent later or be reminded of unanswered messages. ToutApp provides email Templates, analytics, reports and more. 410labs have a couple of different apps to help you work with your email.
Did you ever wish to limit the messages that reach you? OtherInbox has created several solutions to manage email overload, such as automatic foldering. AwayFind allows you to get AWAY from your inbox, letting urgent emails cut through the clutter and FIND you instantly. sanebox separates emails that you must deal with right away from ones that can wait at least a couple of hours.
Want to look at your email in a different way? PhilterIt brings your inbox to life with rich icons. WiseStamp builds an Email Apps Platform that enables users to utilize their emails with social, productive and engaging Email Apps. Unroll.me promises to end unwanted subscriptions and rollup the rest into a personalized and organized overview. Movable Ink allows businesses to stream live content into emails while also providing advanced analytics.
But there’s more! Textbox provides powerful writing tools for gmail (and tumblr). Nveloped makes it easy to send and receive secure email. FollowUp is an email reminder service. Contactually automatically prioritizes and prompts you to take action with your email contacts. And there’s even Maily for your kids!

Ufff! And it’s not just startups! Concepts like Persona or Reinventing Email are being thrown around and getting a lot of attention (just look at the number of comments in HN [1, 2]).

new EmailClient()

Creating something totally new is not only more ambitious but also more risky. The most notorious ones have been shut down. Sparrow was acquired by Google. Fluent mysteriously closed during the Summer. And even Thunderbird got downsized by Mozilla.

But that doesn’t mean others are not going to keep fighting. Postbox’s desktop app for Gmail/Mac OS X has received rave reviews. Riparian Data’s work with big data appears to be very promising. And Mail Pilot had a successful kickstarter project with its mission to reimagine email.

Paul Graham speaks and the tech world listens

In pg’s famous essay Frighteningly Ambitious Startup Ideas, email was a prime target.

“2. Replace Email

Email was not designed to be used the way we use it now. Email is not a messaging protocol. It’s a todo list. Or rather, my inbox is a todo list, and email is the way things get onto it. But it is a disastrously bad todo list.”

And since then, unsurprisingly, there has been a lot of talk to try and “fix email” using the suggested approach.

ActiveInbox has been around for a while, trying to make your inbox your donebox. Previously, Y Combinator had backed Taskforce to help them convert emails into tasks, but the founders have since moved on to Siasto.

There are some new kids on the block: GrexIt, a collaboration system that makes extensive use of gmail’s labels and Taskbox, which turns your inbox into a task list (or as they like to say… a Taskbox).

Even classic task management startups are jumping into the email: Any.DO allows you now to add tasks from Gmail. ~~Orchestra To-do, the App Store's 2011 Productivity App of the Year, is working on a mobile-first email experience from the ground up~~ And we’ve all heard about Mailbox from Orchestra, right?

But how well will they fare against the well-funded Asana, who’s original mission was (apparently) to kill email?

YC is also backing different email startups: Submittable is trying to replace email with a “Dropbox-like solution” while Streak is a CRM in your inbox.

The Big Players: Google, Microsoft, Yahoo, AOL and… Facebook?

Many of these email startups are building on top of Chrome+Gmail, taking advantage of the OAuth access and targeting the 400+ million user base, many of whom are SMBs using Gmail Apps. Development on top of this platform is strangely popular, considering the huge amount of work it involves. But it seems like a trend that’s going to continue. And expect more from Google, at least according to the words of the CEO: “I could probably give you a list of 10 major things that are wrong with email”.

Microsoft has not seen many startups target its most popular webmail service, Hotmail. Will things change with the new Outlook.com? On the other hand, the desktop counterpart Microsoft Outlook has historically provided an API for “add-ins”. The new Outlook 2013 comes with Office’s new web extension platform, “Agave”, which will allow developers to better target what is probably the most popular email client around.

The previously most popular webmail service, Yahoo Mail, has also not seen much interest from email startups. They’ve presented a redesign of their services at the end of 2012, but with Gmail taking the crown, will that be enough? What will Marissa do?

AOL has been busy with its email service. In late 2010 Project Phoenix was revealed but less than a year later development was halted. It seems like efforts are now concentrated in Alto, a project presented in the second half of 2012.

Facebook’s first move into the email world was through the notorious Project Titan. But what was presented at the time fell short of the expectations. Since then, its messaging service became to resemble more and more an email client, with attachments, @facebook.com addresses and a two panel layout similar to Apple Mail. And during last summer, it switched everyone’s profile to a public @facebook email address. Clearly there are more things up their sleeves.

Powering the backend (also known as “The Arms Dealers”)

Email innovation is not limited to user-facing apps. Action is also happening behind the curtains.

Context.IO provides a REST API for email that makes it easy and fast for developers to connect to a wide variety of email accounts and integrate that data in their application. They’ve been acquired by Return Path, the self-called global leader in email intelligence (and also the buyer of OtherInbox).

Rackspace’s Mailgun (also acquired last year) is like an arms dealer for innovators looking to disrupt the email space.

SendGrid provides a cloud-based email infrastructure with reliable delivery, scalability and real-time analytics. They face competition from recently funded Mailjet.

And for helping companies with their email marketing, there’s MailChimp or Constant Contact, among many others.

The world is going mobile, isn’t it?

Email and Mobile used to mean Blackberry. But today, mobile email is pretty much dominated by Android/Google and iOS/Apple.

With everyone pointing to mobile as the next great battlefield, it’s clear that there is a huge opportunity here. The most known disruption attempts were quickly acqu(h)ired (i.e. reMail and Sparrow).

Email startups have been shy and taking a very conservative approach to mobile. Most are making apps that merely signal their presence in the mobile world. And very few have dared to go mobile first. This might be due to the fact that startups that have plugin products don’t have the same freedom in mobile as they do on the browser. They would have to hack into the mobile email clients to be able to reproduce their functionalities and - more important - their visibility. Porting a product into mobile means a separate app, which means not being immediately present when users check email in their phones.

The great exception is of course Mailbox. In just one week they amassed 150k users and +800k reservations! And their path? From a task management app (Orchestra ToDo) into a mix of email client with task management. By building email client functionalities they overcame some of the problems described before and control the user workflow when checking email. Pundits were quick to suggest that someone should buy them (1, 2). Dropbox listened!

What’s tomorrow like?

So what’s in the future for email? Interesting things, no doubt about it.

Originally written on the 26th of April of 2013.

A (not so) Steam-ing Success

Sun, 20 Aug 2017 00:00:00 +0000

Note: This is an old post that still seems relevant. It’s also a better fit for this blog, so I’m reposting it here.

Ah.. Steam. One of those late bloomers.

I recall when Steam was released in 2003 by Valve, forcing every single Counter-Strike player into it. Yells of revolt and “Why Valve? Why?” were heard, people in IRC and forums were all pissed and clamoring for the boycott of Half Life and CS.

Back in those days, it was hard to believe that Steam could succeed. Making people pay for games (and somehow have the large number of underage players get access to a credit card) was gonna be a tough challenge. But Valve kept strong.

Fast forward and what do we see today? Steam looks like a “Great Success”. A lot of big publishers have slowly jumped on board and Steam now sells more than the Valve games. Big names arrive quickly to the system and indie devs are publishing there, providing a huge selection of titles and easy access by users.

As of January 2014, there are over 3000 games available through Steam and 75 million “active” users. In February 2014, it was announced that Steam had surpassed 7.5 million concurrent users. In October 2013, it was estimated that 75% of purchased games downloaded for the PC are through Steam.

Steam has become the de facto place to go to buy games.

Bravo Valve. Well played.

Originally written on the 27th of March of 2014.

Many Roads to Rome, Many Ways of Building a Product

Tue, 18 Apr 2017 00:00:00 +0000

Note: I’ve written before about The Process. My mind came into this again, and while there are some nuances in this post, there’s also some overlap.

The Process

One common theme around companies that have achieved some level of “success” is the attachment to the way of doing things, The Process.

It’s understandable, as it is probably this process that led them to where they are, that made them become “successful”. Trying out something different is risky and prone to failure and no one wants to fall from their ivory tower. If there’s room for improvements, those must be made on top of the existing process, the process itself cannot be suddenly replaced by something completely different.

So here we are, doing things the same way as always. If the project fails, the blame can’t be on the process, for that has always led to success. Were any improvements added? Then maybe that’s your culprit. If not, then the people are the ones responsible for not succeeding - maybe they are inexperienced with the process or maybe they missed some key detail.

And there is a funny invisible repercussion to a process-first company - hiring turns into a game of hiring for “fit”, where fitness isn’t really guided by general intelligence, being a good person, and ability to do the actual work, but mainly about fitting into the process. Culture, they like to call it these days.

I’m obviously being simplistic here, there are many good reasons for companies to default to a process. For one, finding a good process every time a new project starts takes time and effort, something that’s not really abundant. And a bad process has a big impact in the output, so one should play it safe. Totally valid reasons.

But following the same road will at most take you somewhere where you’ve been. That’s ok if that’s your goal, but if you’re into actual innovation or in exploration mode, then you probably want more or something different.

The Processes of Building Products

This post was motived by how companies go on to work on their new projects.

The obvious thinking is that the type of project (critical or experiment) and its goals should help guide how the project should be approached. There are also the usual considerations of how fast you want to launch, how much importance you are giving to technical debt and all of the liking. But the people that are picked to work on the project should have an important say. For example, applying the same process to different individuals most probably leaves a lot of value on the table, because the individuals might be suited to work in a different way.

Having a team of two engineers working “alone” for a month might produce better results than the usual team of five being overseen by a product manager.

As with many things, it’s about your definition of success, what axis are you optimizing for and the trade-offs that you’re willing to make.

Snap, don't Chat.

Wed, 05 Apr 2017 00:00:00 +0000

You’ve seen it before: tech company goes public, the crowd goes wild and everyone’s wondering about its future. But this one might be more intriguing than the previous stories. Hopefully it won’t vanish as fast as its messages.

Ghostface Chillah

You know about Snapchat - the social-network-slash-disappearing-messages-slash-media-posts mobile app that approximately 150M cool kids use on a daily basis. And it has been an interesting evolution to follow: what started as a way to send disappearing photos to your contacts has slowly evolved into a lot more. And it’s the more that I want to focus on.

Snapchat is no longer just about chat. It has effectively become a media consumption and creation app. And perhaps this hints to the general direction of the company (if dropping chat from the name wasn’t clear enough). Yes, the majority of users are probably still using it just for the chat and filters, but there’s a significant user-base that’s just opening it to check out stuff ¹. As it grows, expect people to use it less and less for communication, and more and more for eyeballing - I believe that if they’re not there yet, they will eventually converge towards the 80/20 rule.

Snapchat is actually more like an early days mobile version of Youtube than a media messaging app!

The usage of the camera and features like Stories turn Snapchat into a great media-broadcasting app (just ask Justin Kan). It effectively offers an easy way to create content that can be done by anyone! No need to add a title, description, tags, or anything else to your post - just snap and share!

But Snapchat isn’t limited to content from your friends or people you follow - there’s branded content. And once the product evolves (eventually incorporating long-form content?), we’ll start seeing more music-video debuts and perhaps even original content à lá Netflix!

Business wise, a new-wave digital media (video and interactive content) company makes a lot of sense. TV ad money is flowing into the digital world, no matter what some might be saying.

One big way Snap thinks it can do this is by grabbing TV ad budgets, a shared goal with digital giants like YouTube and Facebook. This money has been slow to move from TV to digital video, but Snap thinks it’s in the best position when that accelerates.

That’s because Snap considers Snapchat ads already superior to those on television, and way better ads than other digital video competitors, according to the Snap S-1 filing. We previously looked at Snap’s argument for why that is, and in the wake of the IPO frenzy, it’s a good time to revisit it.

Snap is in the pole to capture a large share of those dollars. While Snap might be following Twitter’s revenue growth footsteps, Twitter hit a point where they had no guidebook to follow - the product hit limitations on how to make money. Snap won’t have that problem for a while - they can follow TV’s guidebook, a cow that will produce milk for many years to come.

Competition and New Products

The Great TV Ad Dollar Migration hasn’t been unnoticed by Facebook and Google. That’s one of the reasons Facebook/Instagram/WhatsApp (FIW) have been rushing to copy Snapchat. And Youtube is trying out new old things like launching a TV service!

But there’s always this problem with playing copycat and adding to your existing products features that bring in unnatural use-cases versus a product that’s been adding features through “evolution”. The former case is not just making a mess out of their product but there are natural limitations to where it can go (without big redesigns). It’s easy to copy, it’s harder to come up with new ideas that make sense. How the FIW apps will evolve is a big mystery (and the fact that Facebook hasn’t had much success of late with copying is also an added factor of interest).

Snapchat has much less baggage to carry, e.g. it’s still mobile-only AND they only care about high-end devices; contrast this to WhatsApp, who has historically gone the extra mile to make sure they run on every device. In terms of product development this has a huge impact - Snapchat can be lean and try out many different things quickly and release things in no time; the others will at most only be able to roll out features progressively or risk taking too much time.

From the FIW bundle, I can only see Instagram being an actual challenger to Snapchat, and even they will need a big redesign to effectively compete, assuming that’s their priority. If not, they risk irrelevance. It’s easier for Snapchat to invade Instagram’s turf than the other way around.

Youtube is another competitor for the Ad Dollars, but they come from somewhere else. Youtube is the incumbent, the default place for online videos. That’s why it makes sense for them to launch a TV service, it’s Cable TV that they’re thinking about, not ephemeral messaging apps. Youtube was never mobile first and the default use case when opening their mobile apps isn’t to record a video, but to consume videos. And that’s ok, because that’s what most people want to do. And uploading a video follows a process closer to what professionals do, because Youtube’s thinking about “quality” videos, not shorts, with crappy image quality, of people getting lost at sea. And this mentality is their weakness, for they are going for the same treasure chest.

? ²

Snap-volution

In order to reach the promised land, Snapchat needs to change. Quite a lot. My opinion is that they will ultimately need to tear down their walled gardens and hold the interesting content for longer so that it becomes available to anyone, not just followers. An open model, similar to what Twitter has, is actually something to follow. Content is king, and that should be their focus - helping users create good and find good content. That might mean little things like more filters and lenses or trending snaps but other outrageous ideas that Evan Spiegel & co. need to come up with.

The good thing is that they seem to understand this and working towards it. And being based in LA, the capital of entertainment, probably won’t hurt.

Snap has the essentials to go and change how we create and consume entertainment. Let’s see how this goes.

Snapchat Content Survey: How Much Millennials Actually Use Live Stories, Discover and More ↩
There’s also this giant lurking in the shadows - they’re very good observers, patient and very effective. And they’re also straying into online video, from many different directions. I’m talking of Amazon, of course. It’s not enough that they operate the largest livestreaming platform and are into video entertainment, but they’re also experimenting with regular TV content. The missing piece in the puzzle? Mobile based user generated content. While it’s a stretch that they would go into this market, it’s always something to keep in mind. ↩

An Ode to Google Keep

Tue, 27 Dec 2016 00:00:00 +0000

Here’s a cheer to a great product: Google Keep.

Google sometimes reminds us they can actually build good products and Keep is one of those reminders. Launched in 2013, Keep is a simple note taking/organizing app. It is an alternative to Evernote, Microsoft’s OneNote and what Dropbox Paper would’ve liked to become (sorry, cheap shot, I know).

“A note taking app, how hard could it be to make something good and successful” you ask? Truth is, it’s never easy to create and maintain multi-platform apps (web, android, ios), have hundreds of millions of users (even with the help of the Google brand and Android integration) and get good reviews. Just ask Evernote and their recent string of screw ups.

Keep has managed to keep (heh) the UI minimal but powerful, and the apps (generally) bug-free. But the most important thing: the core functionalities work very well.

Hopefully Keep will continue to follow its current product philosophy.

Matt Berninger on Startups

Fri, 28 Oct 2016 00:00:00 +0000

Matt Berninger, from The National (and El Vy) will give you the absolute best advice on startups.

What?

Really?

Yeah.

Don’t know who The National are? That’s ok, Wikipedia can help you.

The National have had a very interesting path. They have become successful and their (overnight success - not!) story is familiar to anyone who follows the tech industry:

The National’s rise to success has been a rocky path to say the least. The band managed to put out four highly acclaimed albums throughout the ‘00s without ever really gaining a level of popularity they deserved. That all changed in 2010 when they dropped their fifth studio album High Violet, whose monochrome melodrama saw the band pushed to the forefront of modern rock.

On the early beginnings and having no users

Something I genuinely took away from the film was a quote from you about not letting bad luck sink you into a spiral way, and you reflected that things don’t always go right for you either. You’re quite open about those first six years of the band’s career being a rocky struggle, I was wondering if you could perhaps think of some moments early on for The National that were a bit like the projector failing at Tom’s opening screening.

Oh man, we had so many from just the embarrassingly simple ones like showing up to a gig and literally no one being there. We had several of those gigs where we’d play for completely empty rooms and in one case the bartender just said, “Hey listen, I’ll still pay you even if you don’t play so we can all go home.” And we were happy to do that, we got paid for playing and we didn’t even plug in the guitars. Actually, we set up and we were about to play and still no one was there, even during the sound checking, and the doors had been opened for hours so it wasn’t like the time was wrong. It’s just that nobody came to see us, zero people. That happened probably about four or five shows, zero people came. There was one show in Akron, Ohio, where there was only one person and it was Patrick Carney from The Black Keys. He was the singular, only person that stood in front of our band and watched us play. It was so funny because their band was a tiny little band still at that point and I remember recognizing him and thinking “I think I’ve seen that guys face on Pitchfork or something”. We had years of that.

On being overshadowed by the “competition”:

We had so many little bumps like that. Another thing, for example – and I think this has been documented a little bit – when we put out Boxer or Alligator, one of our friend’s band who we love, and we still love the band and love the record, was Clap Your Hands Say Yeah. We invited them to open our tour and their record just exploded into this international hit phenomenon and suddenly all of our shows sold out almost overnight when they hit. So we went out on tour with Clap Your Hands Say Yeah opening for us, and the minute they left the stage, half – and some nights more than half the people in the room – left. That was a very character-building experience for our band. To tell you the truth that hurt [laughs]. It wasn’t their fault, they made a brilliant record [2005’s Clap Your Hands Say Yeah ] that people were dying to come out and see, and I just guess at that time people weren’t that interested in our band or I guess our fans weren’t buying tickets fast enough to get in [laughs].

On domain names:

When the band was formed in 1999, it was called “The National”, although the domain name of the band’s website is americanmary.com because, according to Matt Berninger, “[i]t’s a song off our first record. We never thought of changing the (website) name, although we should have.”

On leaving your +$100k job to start your company and then eating ramen all day:

We slept in youth hostels and vans, and I remember once when we slept in a youth hostel where I went to the bathroom and when I came to my bunk, some other man had put his wet underwear on my bed, on my pillow. At that time I’m in my mid-30s. I’m 32, 33 and I had left a really good job to do this, and I’m laying in a hostel with drunk Scottish kids, who are puking and screaming, and international travellers are putting their wet underwear on my bed. I’m a grown man and I remember lying there nearly in tears just like “What am I doing with my life?” I can’t tell you how many times I’ve stared at the roof of a van, or the ceiling in the middle of a cold bed at night on somebody’s floor thinking “What am I doing with my life? This is not worth it”, and then it slowly turned around.

On thinking about quitting:

We seriously discuss that every few months, every time we make a record, every time we finish a tour. When I said that I was asked “Do you ever think of the band ending?” and I said, “Yeah I think everyone thinks about the band ending all of the time.” There are those dark nights when you haven’t seen your kid or your wife in four weeks and it’s cold and you’re sleeping on a bus in Germany. Even when things are going amazing, you’re still living so much of your life on a bus away from your families. It’s not something we want to do forever; we don’t want to be life long touring musicians. We don’t want that to be our lives, so it’s something the five of us talk about a lot.

On finding that life/work balance:

Now that some of the other guys have kids it’s an easier thing to talk about because we’re all on the same page of like “What’s the best way to do this without losing our minds, but still pursuing it in a healthy way?” We’ll still be making records forever, but we’ll slow down on that [touring] for sure. I think every band has moments several times a month where they’re like “I don’t want to do this anymore, it’s so stressful” so we talk about it a lot, but no more now than we have before. We’re definitely in a healthier place than we were when we made Boxer.

On pivoting and “dropping out” of college:

When I went to college, I decided I was going to be a doctor. I was in pre-med for one year at Miami University of Ohio. I hated it. I wasn’t bad at it, but I hated it. The next year, I switched to sculpture. Pre-med to sculpture is a pretty big leap. Then I realized that maybe being a sculptor wasn’t going to be enough. I didn’t love it enough. So I quickly pivoted to graphic design. (…) My point is, early on, I got used to bailing out on one life and trying something else. Not bailing, but pivoting. I was comfortable with doing a significant pivot from high school, to pre-med, to sculpture, to graphic design, and then to New York. For whatever reason, I was never nervous about jumping into the unknown. I was lucky that way.

A reminder about the privilege of having a supporting family:

It’s not like my parents had wanderlust, but they were very encouraging. They wanted me to be happy—that’s what success meant to them. When I told them I was switching from pre-med to sculpture, I had this sense they were going to think it was great. Most parents would probably say, “What?” But my parents were like, “If that’s what’s going to make you happy, that’s awesome.”

On the importance of being early:

We got jobs at places that were just starting to do websites and new media. It was at the beginning of that. We got jobs that were kind of well-paying for just out of college. Then, because they were well-paying, and no one else was doing it, we rose in the ranks. I went from Junior Designer to Creative Director in a handful of years. I felt professionally motivated and excited. I felt like, “Wow. I can be a grownup man in the world, pay my own rent, and buy my own TV.”

On why you should be in the Bay Area/Tech Hub City:

Proximity to other people having crazy, reckless, delusional pipe dreams is one of the most motivating things. That’s why places like New York are so rich. But yeah, we started slowly. It took a long time to go from playing as the first band at Mercury Lounge at 6:00 p.m. to headlining. It took us a few years to get anywhere close to the world that Interpol were in… at least it felt like that.

On savoring the little wins and starting small:

Things got better very slowly and incrementally for us, and we could taste it. All of our tiny victories - moving from the 6:00 slot to the 7:00 slot to the 8:00 slot - were significant to us. The fact that it took us a while to get to that 10:00 p.m. headlining slot at Mercury Lounge made it all the sweeter. Being in the shadows gave us time to find the weird branches that we’d end up going out onto. The fact that there wasn’t much light on us for the first few years helped us figure out who we were.

On overnight successes:

A lot of people think they have to be a genius the minute they walk into the light because they think that’s what Bob Dylan was. Even the artists you think are perfect artists, probably had more bad output than good. I’d almost make an argument that Bob Dylan has more bad songs than good songs. But his good ones are the best songs in the world. I think people put too much pressure on themselves, and they can’t handle it.

A reminder about the privilege of having a safety net:

You get out of college and the big question is, “Will I live?” Like, “Am I going to be able to survive on my own?” Every person goes through that phase where, even though you might not live at home, your parents are still helping you. Not everybody, but lucky people, and everybody in The National, I think you could say that we were lucky… None of us were worrying about eating, although many musicians are.

If I had to, I could go back to a job. If I had to, I know how to do an interview. I know how to get a job. If this band fails, I can always go back and be a Creative Director or do lots of different things because I’m good at getting jobs. I’m good at ingratiating myself and making myself appear useful. I knew I could have a job—maybe that’s what it was—it wasn’t the job itself. I figured out I can come back if this rock and roll delusion turns out to be a delusion.

On being afraid of putting your work out to the world:

The funny thing is, I’ve lost a significant amount of my anxiety about what I’m making. I still obsess over it, but I’m not as nervous about making a fool of myself. I used to be very nervous about humiliating myself. I’m not sure that’s such a good thing, but it’s liberated me in a lot of ways. You’ve got to be fearless, but you also have to respect that every thought you have while falling asleep isn’t necessarily worth publishing.

On the importance of believing in yourself and being delusional:

I do think it’s important to be a little bit delusional to be able to do any kind of art. You have to believe in yourself more than everyone else does because, don’t worry, the people that don’t believe in you are going to be very vocal. You have to have a very healthy delusional ego to do anything. But at some point, when your ego starts saying, “Everything I do is awesome,” then you start putting out a lot of crap, and it’s not all awesome. It’s a fun, interesting, dangerous place to be as an artist.

On the importance of a team:

I will say that I don’t remember there ever being a time where we’re together in the studio working on something where at least two-thirds of the people in the band weren’t super excited about it. There’s never been a time where as a band we’ve all looked around like, “Ugh, why are we doing this?” Never. Never even close. I think we’re lucky. Mostly it’s because when somebody gets exhausted or full of resentment or runs out of ideas, somebody else comes in and gets everybody excited again with a different thing.

We’ve got a lot of ambitious people that know when to pick up the baton when somebody else is running out of steam. You need time to get away from shit and come back to it fresh. Bryan is a brilliant drummer. Sometimes he’s feeling it, and some days he’s not feeling it. What he does is so mental and physical and all this kind of stuff, we’ve learned to let Bryan do what he needs to do to get himself either mentally or physically into a space where he can kill it. He’s a drummer that doesn’t sound like other drummers and we give him time to be that drummer. He would maybe argue with us.

That applies to all of us in different ways. Those guys know that if I feel like I’m not happy with the lyrics yet, it’s worth waiting. They want me to be happy. Over time, we’ve learned to respect each other’s weird, sometimes clashing, skills, sometimes clashing personalities. We’ve definitely learned to try to avoid the things that exhaust us and lean towards the things that remind us how we dance well together.

On learning how to take punches:

We had lean years. We learned to have fun with it. Our skin got thicker. We also just learned, “Tonight sucked. Nobody was there in Philadelphia, but tomorrow is Indianapolis. You know what? There were five people in Indianapolis. That’s better than no one.” We actually were able to tell ourselves that five people at a show was good, and it was, because it was better than zero.

And to ignore those Techcrunch praises:

Those things definitely gave us perspective. I do think that if you get a big, glossy photo of yourself in a magazine too soon, you think you’re there already. When you realize that, “Oh, shit. The next photo might not be bigger. It might not be glossier. Our name on the poster at the festival isn’t getting bigger. It’s actually getting smaller.” that could be crushing. Some people can’t handle it. We had a lot of phases where the type size of The National on a poster at any given place was big sometimes, and then it would go back. We were constantly reminded that we were nobody.

And doing what it takes to pay the bills:

We probably wouldn’t be a band if we hadn’t had a song in a Saturn commercial way back then. This was way, way back. They told us it was a hybrid [car], and it turned out not to be a hybrid… but we were able to pay off bills. Now there are so many television shows. They don’t pay the way they used to, but there are a lot out there. I think there are ways that musicians can keep the lights on and pay the bills without being a big band. That’s a whole other conversation, the ethics behind letting your music be used in a product placement is a whole other debate.

And if you’re getting investors, being careful who you partner with:

With major labels, we were lucky we always went with the label we felt wasn’t going to fuck us, and they haven’t, because they don’t depend on us to sell millions of records. We don’t make videos that cost hundreds of thousands of dollars that they have to recoup. We went with a label that was going to be patient with us and they followed through on that promise. They’ve been patient with us, and everybody’s doing well. We’re doing well, they’re doing well. We’re happy. We were smart enough to be careful and protect our ability to make the type of stuff the way that we wanted to make it.

On things taking time, and failure. And bad reviews:

Patience… and respect that it takes time. Make sure to give yourself that time. Also respect the failure. Respect every time you play the show when nobody is there. You learn how to be better. Respect that terrible, really bad review. Never agree with them, but respect them. My other thing is when you get reviews, read them all, but read the good ones over and over. Just lean that way. Believe in those, but let the others go, because they’re both probably right. Believing in the bad reviews doesn’t really actually help you much. You might learn a little something or other, but have delusional confidence. Don’t expect it to be there right away, and don’t expect it to stay, but you’re in control of it. You can make whatever art you want to make, and if it’s good, people will love it.

On releasing too early and no one using your app:

Just trust yourself and like the stuff you’re making. Don’t put it out until it’s cooked, but then also don’t be crushed if everyone hates it at first. People will provide you with far more reasons to stop and to give up than to keep going. Obstacles and defeats will be a part of it from the second you start. Humiliation and defeat is an essential part of being an artist.

On unicorns:

If you’re doing anything artsy, the rest of the world’s job is to say whether it’s any good or not. Because you’re making art. You’re trying to be a magician. You’re trying to be an illusionist. Literally, you’re trying to create a unicorn, a good song. If you put out something that’s like just a horn taped to a goat, people are going to say, “Nah, that’s not a fucking unicorn, man,” but occasionally somebody can make a unicorn, and it’s like, “Holy shit.” It makes you believe. You’re making art. People are going to tell you you suck most of the time, but a couple of other people might be like, “I don’t know man. That might be a fucking unicorn. I kind of like it.” The truth is, everything is just a goat with a horn taped to it, but sometimes that’s fucking even cooler than a unicorn.

Mental toughness and delusion, again:

Those are the weird mind games you have to play with yourself. You have to go to bed and tell yourself it’s all going to work out, every night, one way or another, because then it will. If you don’t tell yourself that, it probably won’t. Delusion is very healthy—up to a point.

Here are the complete articles:

http://www.fasterlouder.com.au/features/36709/The-Nationals-Matt-Berninger-Its-hard-to-sing-when-somebody-grabs-your-crotch

and

https://thecreativeindependent.com/people/matt-berninger-on-patience/

Now he just needs to apply his persistence to surfing. 😂

JAWS 6 TRAILER. #surfing

A video posted by EL VY (@el__vy) on Oct 2, 2015 at 5:15am PDT

The Institutionalization of Process

Mon, 24 Oct 2016 00:00:00 +0000

AKA when previously agile and innovative startups become slow and fossilizied companies that stop releasing new things.

You know how the successful “Valley” story goes: new consumer startup is red hot, attracts millions of users (and funding dollars), releases new features every month that get covered by Techcrunch, and it eventually goes public. Then the innovation ¹ cycle starts to take longer and longer until it becomes pretty much non-existent.

(An example, as of October 2016, of a startup that’s following the upward part of the “Valley” dream? Think Snapchat.)

Once those companies become public, what they usually try to do in order to continue to innovate is to acquire interesting startups so they can inject their organization with new innovative blood. Think Google with Youtube and Android, or Facebook with Instagram and WhatsApp.

But if they fail to innovate with the new blood they might end up like the current sad-stories of these days: Yahoo and Twitter.

So why does this happen? Here’s an interesting hypothesis:

(…) it’s that people get confused; companies get confused. When they start getting bigger, they wanna replicate their initial success.

And a lot of them think “Well, somehow, there’s some magic in the process of how that success was created.” So they start to try to institutionalize process across the company.

And before very long, people get very confused that the process is the content.

(…)

We had a lot of people who were great at management process. They just didn’t have a clue as to the content.

And in my career, I found that the best people, you know, are the ones that really understand the content.

And they’re a pain in the butt to manage, you know? But you put up with it because they’re so great at the content.

And that’s what makes great products. It’s not process, it’s content.

Food for thought, eat it at your own discretion.

Simply defined as releasing new “successful” things. ↩

The Promise of GitHub

Mon, 26 Sep 2016 00:00:00 +0000

I initially started writing this post 3 years ago. But then I stuck it in the burner. Now it seems like a good time to bake it.

Lets git it going, shall we?

For those who don’t know about GitHub, it is the best place to share code with friends, co-workers, classmates, and complete strangers. Translation: they provide source code management tools. This space was previously dominated by Sourceforge and Google Code, who did not adapt to the new times. Today, it faces competition mainly from the publicly listed Atlassian and the upstart GitLab.

GitHub has become the de facto place to freely store open source software projects. And it has also seen huge popularity for its paid plans among tech companies and individual software developers. That is why, when it went into fundraising mode, it was able to raise a total of $350M ¹ - $100M in 2012 and then $250 in 2015.

While it is a good polished product, most of it has been the same for a long time, with only tangential improvements, and there has been a feeling that a huge margin for improvement exists ².
Given the massive funding and headcount (almost 500 employees at the beginning of 2016), why isn’t GitHub perfecting their product and expanding into other areas?

GitHub, deconstructed

Succintly, these are GitHub’s main products:

GitHub source code hosting and management: provides repositories for source code with all the attached goodies such as issue tracking (a basic issue management system), collaboration tools (contributors and pull requests), documentation platform (wiki), statistics, integrations with third party services, and some more;
GitHub Gist allows you to instantly share code, notes, and snippets;
GitHub Pages lets you host websites for you and your projects directly from your GitHub repository;
GitHub Enterprise is the on-premises version of GitHub, which you can deploy and manage in your own, secure environment

As mentioned before, GitHub has been criticized for its slow speed of execution and not improving its core product very much. Given the context of the imense fundraising, the criticisms do seem valid. But lets try and dig deeper to back them up.

I looked at GitHub’s Blog, more specifically it’s “New Features” category (where new features are announced), and then counted & grouped the pages by year ³. This can be seen as a rough indicator of GitHub’s internal activity.

Year	Num Pages	Num Users
2012	4	1.7M users
2013	6	3M users
2014	4	8M users
2015	2.5	10M users
2016 (so far)	2.5	15M users

The results seem to match the sentiment that has been expressed. It looks like that in the past 2-3 years there has been a slowdown in the number of new features announced. Yet this hasn’t stopped its immense growth - some vocal users complain, but things seem ok. But are they?

What’s been going on?

2016 has been eventful for GitHub. The year started with the Dear Github open letter ², which attracted a flood of comments from the software developer world ⁴ supporting the cause. A few weeks later, repercussions of the management and culture change (first started in 2014, after the resignation of the then-CEO ⁵) started being felt - employee departures and disband of the flat structure and remote work ⁶.

Perhaps GitHub has finally heard the cries from its users and started doing something about the core product, with the recent announcement of new tools, forums, and features. But that brings us into to the land of how-things-are-done, and not everyone agrees on how they’re doing it.

Brand New, But Just The Same?

This is the example of how hard it can be for companies, especially successful companies, to alter their course even slightly.

Github success was about making code repositories pretty and social. But this is no longer enough to keep evolving as a company of their size. Yet the internal product DNA still revolves around repos, so as a result, issues, wikis and now Projects (!) live under a single repository.

This is backward.

Source: The top comment in HN regarding the new GitHub features announcement.

GitHub, in some aspects, feels like a lost cause. It remains to be seen if they’ll be able to get their mojo back or if it’s going to be a long slow decline. Given their warchest of hundreds of millions, it would seem unwise to bet against them, but on the other hand, the pattern has been observed with other companies in the same position.

The title from their homepage “How people build software” - is what GitHub could’ve become. But it hasn’t, and there are no signs that it will be. Instead they might now be facing increased competition, specially with such a low barrier for users to switch services.

The Shadow of Atlas(sian)

GitHub’s bet in the Enterprise market faces strong competition from Atlassian, who IPO’ed at the end of 2015. After an initial decline in the stock-price at the beginning of the year, it has now gone back close to its previous heights.

In terms of product, while JIRA is the king of project management, it is not supreme. And the once reigning Hipchat has taken quite a hit from Slack. But Atlassian products are still solid and it has had over a decade to ingrain itself in the enterprise world.

While Atlassian has never stuck out as the flashing innovator that the Bay Area adores, it has always looked like a consistent fighter. And still flush with the IPO money, who knows what they might buy.

GitLab

GitHub’s mightiest competitor might simply be another startup: GitLab. Backed by YC and with a fresh $20M series B, GitLab has a clearly defined vision ⁷ - that I always thought GitHub should be doing - which is executing at a rapid pace.

They have understood that they should target the development pipeline, i.e. how people build software, and make the tools for that. It is to be seen if GitLab’s strategy is the right one, but the fact that the CEO and others are listening and talking with their users bodes well for them.

You know, kinda like what GitHub’s CEO used to do.

GitHub on Crunchbase ↩
An open letter to GitHub from the maintainers of open source projects ↩ ↩²
I decided to start counting in 2012 because that’s when funding came in. ↩
Dear GitHub ↩
A Beloved Silicon Valley Startup Is Under Fire For The Way A Female Employee Was Treated — The GitHub Fiasco, Explained ↩
GitHub is undergoing a full-blown overhaul as execs and employees depart — and we have the full inside story ↩
GitLab Master Plan ↩