43 comments

[ 3.2 ms ] story [ 78.1 ms ] thread
I am not sure what to make of that list

* It mostly lists generic software (Google-docs?) * It's extremely MAC centric

Start working at home with Google! It’s by-far the best job Ive had. Last Monday I got a new Alfa Romeo from bringing in $7778. I started this 9 months ago and practically straight away started making more than $83 per hour. I work through this link,>>>>>>>> www.jobs34.ℂom
>.ℂom

that's a new one

At $7K, I'd say that Alpha isn't a new one.
I'd like to think of it as "Tools we love to use"; or "Tools I like so much, that I want to tell others about it".

Clearly Google docs solves the problem of versioning and sharing quite well and that probably resonates with a lot of folks in stark contrast to how they have to share a excel or word document and struggled to keep the versions updated.

Dunno what anyone else thinks...... Its just that describing google-docs as software jarred with me. I guess I see google-docs as a "website" or a "service", not "software", as such.

Why?

Not actually sure. I think I see software as something I buy and install on a local machine. But my, possibly incorrect, gut reaction is that services on websites are not "software".

> Almost everyone uses a Mac. Where? I, actually, don't know anybody who uses mac.
It depends on which developer groups they used. It seems certain web languages attract people who use Macs, where this data probably came from.
It also includes designers which in my experience are almost 100% mac users. So that will sway the stats more in that direction as well.

What is weird to me, is under Software we see Windows 7 and even Windows XP but we don't see any form of linux.

and we see a lot of people using Macs but running Windows on them.
I used to see that a lot in China 5 years ago when mac hardware was becoming popular. Then about 3 years something happened: they stopped installing pirated XP on mac hardware and rather were already used to OS X because of their iPhones/iPads. Quite strange.
True that. The only Mac user in my team of about 100 developers & devops folks is our designer. The rest are split between Linux (Arch & Ubuntu) and Windows 7. The other major dev group in my company (also ~100 people) has 0 Mac users and is also split between Windows & Linux. The .Net guys use Windows and nearly everyone else (Java/Python/Perl/DBAs/sysadmins/PostgreSQL/MySQL) prefers Linux.
>> "What is weird to me, is under Software we see Windows 7 and even Windows XP but we don't see any form of linux."

It could be that the interviewees used several different distros. So Linux would have made the list but splitting it into distros makes the numbers for each too insignificant to be in the top 10.

That also means you don't know anyone who writes iPhone app, and I don't see how that's even possible for a developer.
iOS development necessitates OSX.
Notepad++ is more popular than both vim and emacs. I don't believe this.
On Windows it's the best choice really. It's a lot faster than ST2 and there is no interface clash as with emacs and vim.
It's not particularly surprising when you think about it. Users of stenography keyboards are far more "productive" than users of QWERTY keyboards. But everyone knows how to use the latter at some level of competence, while the latter requires extensive training in specific skills. And really, do you need to be able to type 200 WPM?

In the same sense, vi or emacs might make you more productive in the limited sense of typing code, but only after you've taken the non-trivial amount of time required to obtain a level of proficiency with them. And even then, being able to type and edit code more quickly than someone using a more traditional text editor is only a small fraction of what constitutes performance for a developer or designer.

I am the author of the post and happy to answer any questions.
Was Linux represented in your test group?

Was k-mcgrady right that all the different distros made none of them make the top 10 list?

With the people I know and work with Linux seems to be the most popular option for programming so I am curious as to what your statistics show, and why Linux doesn't appear.

- I did not create a test group. I did a statistical analysis of all the interviews on The Setup: http://usesthis.com/

- Among this group, the number of people who were tagged linux was about a fourth: https://github.com/becomingGuru/usesthis_analysis/blob/maste... I did not do a distro distribution. That is good feedback. I will find out.

Since designers hardly use linux, and since half the people are developers, it may be safe to assume half the developers interviewed by The Setup use Linux.

I call sampling bias, if only because very few of the 'influential' developers I know use a Macintosh. I guess that the sample only includes 'hip' independents. Most people in a corporate setting do not choose their workstation. Then, anyone who hosts anything has a heavy dose of his Linux distribution of choice in the mix. Not defining 'influential' doesn't help with the site's credibility.
By the number of times you quote adjectives in your comment it's clear you have some sort of reflexive disdain for the whole concept. However I challenge you to define "influential" in any more objective way than what The Setup does. They interview a pretty diverse set of individuals, but of course they are all somewhat Internet famous for some public work they've done. I think it's pretty safe definition to say that people in the corporate world who don't choose their own workstation and are not publicly known are not as influential as interviewees on The Setup regardless of how smart, talented or accomplished they are.

If you don't like the content of The Setup then don't read it, but this is a simple statistical analysis of something which exists and many people value in its own right. You can take it or leave it, but please don't set up a strawman about how it's not statistically valid when in fact there is no way any such endeavour could be.

(comment deleted)
"please don't set up a strawman about how it's not statistically valid when in fact there is no way any such endeavour could be."

So you agree that it's not statistically valid. Therefore it's not a strawman argument.

Do you know what a strawman argument is?
our company let's you choose between macs and pcs.

quite necessary as we develop mobile apps - just how would you develop for iOS without a mac?

we have a lot of ubuntu and win7 running on those macs, win8 now necessitates new hardware due to touch.

  > Popular hardware: 
  > Keyboard, less than 50%
  > iPhone 4, 60%
  > Apple's incredibly overpriced cinema display, 75%
  
  > Popular softeare:
  > Chrome, only web browser listed, roughly in line with winxp
  > winxp + win7 more than mac osx, yet most people using a mac?
  > google-reader more than mail ?

There is something incredibly wrong with the data used by this article.
The data just represents the population it was taken from. The problem is the inference the author tries to imply.
Hey, I am the author of the post. Here are some clarifications:

- None of the values are the %. They are the actual number of people that said they use that ware. About 70 people said they use iphone 4 and about 40 people said they use the apple keyboard.

- "keyboard" is a slug used for the Apple Keyboard. Half the people that were interviewed wanted to explicitly mention that they use that keyboard.

- No one provided the interviewees with a set of names to tick if they used that hardware or software. In that case you'd have a comprehensive set of user-wares from which you could get the accurate info. I'd detest such an interview. I can not imagine why all of these people wouldn't.

- What we have is a free form text where in the interviewee says whatever he's allowed to. When he says he uses a Mac, there is no reason why he'd also say he uses osx. - That would be redundant.

- If people are asked what they like to consume, no one mentions water or cereals or vegetables, even though everyone has it. What is mentioned is the particular dish or a cuisine that they actually want you to know that makes them feel better. Likewise here, google reader actually helps a lot of these people and they want to share it with the readers.

- It is much more what people like themselves to be associated with. Seth Godin has written a lot about it. People say Google reader much more than mail because they want to be associated with using it. "mail" here, btw is the slug to the actual Apple-Mail client.

Hope it helps set a perspective on how this should be read.

oh... hardware and software we "should" use according to some trends? Apple? Google? what happened to Hacker culture?
“When it comes to computers, what hackers are doing now, everyone will be doing in ten years. Almost all technology, from Unix to bitmapped displays to the Web, became popular first within CS departments and research labs, and gradually spread to the rest of the world.” - PG

I am saying you should rather already be using this now, than doing it later, given you'd be doing it anyway. (from PG, above.)

I think you're incorrect to say that. You're recommendation makes no sense. You're saying 'hackers' should do what a mixed bag of 'developers' and 'designers' are doing now, because it's inevitable they will be anyway in ten years time, because 'PG' said so.

Picture it, self proclaimed 'hackers' all aping each other. All uniform, non distinct from the general population. What would then make them predictors of the future? There would be absolutely nothing unique about what they were doing.

I'm right in assuming at least half of these are designers and not developers. It's particularly doubtful that 'hackers' are going to swarm on to apple hardware over the next ten years given the mac tax and apple's propensity to lock it down.

The "should" use was meant for the general populace, those that adopt something, after the tipping point in a bell curve, to give a sneak peak into what hackers use today. - Not to hackers, who can themselves choose what they want and who form the initial part of the bell curve.
The statistics of this article are quite wrong. Only 2 people using vim? I've read multiple dozen articles of people using vim on thesetup.
The analysis is entirely available here: http://nbviewer.ipython.org/urls/raw.github.com/becomingGuru...

Perhaps there are other slugs "gvim" and the like, which also have a 2 each users.

I will verify this and let you know. Meanwhile pointing to any 3 vim users would help me debug this.

Model M keyboard only at 5%? That sounds surprisingly small. Though open offices might explain that.
Thats actually 5 people of the interviewed-analysed 373. (And not 5%)
Harsh then. I can't imagine programming without my Model M. It also let's my office mates know I'm working...hehe.
This is just a list of basically interchangeable hardware/software products. How much does it matter that certain demographics prefer macbook pro to macbook air or sublime to emacs?

The PG quote is talking about things like the WWW or bitmap displays, stuff that is fundamentally different technology, not just measuring the popularity of laptop brands.