lusr

Agreed; the title of this post is horribly inaccurate and the article is fluff that has been repeated over and over every year in every magazine and newspaper. I don't understand why this is receiving any attention at all; it seems most of the comments here are purely anecdotal data.

According to the article the valuation is $750m, that means they probably purchased 12.5% for $100m. To generate a 2% return on $100m @ 12.5% implies $16m of distributable profits. Assuming 40% of GitHub is net profit (a generous estimate), this implies $40m in revenue.

That represents 476k 'micro' subscribers in one year, 277k 'small' subscribers, 151k 'medium' subscribers. Cheapest business plan (bronze) would require 133k; most expensive (platinum) would require 16.7k subscriptions. Github announced 1 million users on 21 September 2011. I can't find any details about current #s or the % of paying users.

Remember that all these numbers double if GitHub only has 20% net profit (bear in mind this would be 2.5% higher than the insurance industry, which is an extraordinarily predictable and profitable industry, and oil/gas, the most profitable industry, sits at 25%).

So while not 'spectacular' it's not at all obvious whether they can generate the sales required to generate the return necessary, and the optimistic attitude of people posting here without any numbers to support their conclusions is a bit disturbing (not a reflection on parent's comment, but in general).

Looking at the numbers I'll make the following observations:

- the total market size of potential GitHub users has a fairly low limit (somewhere in the 10s of millions I would guess)

- GitHub has some popularist advantages but ultimately competes on price as far as I can tell (I've started moving to Bitbucket because I see no point in paying for something I can get for free, and the recent hack shows GitHub doesn't offer better security AFAIT); as soon as they gain a larger market share there's no reason to believe competitors won't want in on the action, and as GP points out, switching repository hosts is really easy

- while it looks like they'll be able to generate that 2% return over the next few years I'm struggling to see them generating a much more interesting return based on the numbers I'm looking at

- I have a feeling people are confusing feelings of "GitHub is cool and I like it" for implying that "GitHub should do well as a profit generating business for its investors"; they are also probably fantasizing about having $100m invested in their own startups and taking criticism as personal attacks on this 'plan'

GitHub recently announced Subversion bridging:

https://github.com/blog/1178-collaborating-on-github-with-su...

Good point; a physical key is just as good as a password and has the benefit that people will treat it properly.

What's wrong a simple "Okay sir, before you can drive away with your new car you need to pick a password. And before anybody can service the car they'll need your password so please don't forget it, but if you do you can always provide proof of ownership to your nearest dealer and they'll help you reset your password."? That way non-franchise garages can still do repairs, as well.

You're doing exactly what JackC is talking about. You're deciding your way of enjoying and loving music is the only way and anybody who does it in a different way is an imposter. Ironically, by my definition, that makes you a hipster :))

A fair point but (a) my observations are still accurate (just scratch 'client') and (b) going off-topic, it makes me question the management structure at Facebook (re: recent post on HN) where they end up building a complicated set of features but can't build the back-end to support it.

I'm not sure what you're trying to say. I said it's one of the most buggy web applications I've ever seen. Maybe your experience differs, but I cannot see how that invalidates my experience.

I don't really understand your point about complexity. It seems contradictory. You're saying other sites have most of these issues but you're also saying Facebook is much more complex without having much more issues. That doesn't make sense, especially since you're also saying you think these issues are common to what you say is a fragile technology stack under these loads (AJAX) and suggesting that it's okay to then add extra "gadgets and clickables" to an already fragile stack.

With respect to your claims about Gmail, my 2 year old Gmail account currently consumes 815MB of storage across ~6K emails. I use Google Docs regularly. I have never seen the issues I listed w.r.t. Facebook on Google services. GTalk works perfectly across multiple platforms simultaneously. I even find www.gmail.com (mobile & desktop) to work better and more reliably than Apple's Mail application on my iPhone. I have never failed to attach a file to a Gmail email; I've also never failed to upload an image to Docs. Facebook's application - mobile iOS, under Safari on iOS, and on Windows - is consistently buggy.

You seem pretty heavily invested in defending Facebook in such a way that you take my observations as personal attacks. Are you currently employed there or have you been employed there by any chance? Or are you a non-native English speaker?

Facebook's client is one of the most buggy pieces of web software I have ever seen, whether on mobile (iOS) or desktop (Windows - Chrome/IE/Firefox):

* pages frequently don't refresh correctly (content is missing)

* changes to your profile take hours to reflect or reflect inconsistently

* stale notifications persist despite reloading (CTRL+F5) a page

* notifications made within the space of 12 hours are listed as a combined notification

* image uploads fail with no way to recover (dialog just sits there) or often take me to an error page (with the album created)

* the timeline places items in weird locations with no way to fix it (changing the date doesn't work consistently)

* chat windows show inconsistent messages between two tabs on the same browser

That's just off the top of my head. Some of these seem to have been fixed, but I'm always running into new problems. (Note: I'm not suggesting it's easy building a site like Facebook.)

Just because you cannot set all variables outside the body to equal does not mean you should have a free-for-all.

Physical inequalities - not mechanical.

I wonder what the limits are here in the long run.

Will we be able to print cooked steaks or will cooking still be required?

Will we be able to print a cooked steak, with off the shelf materials from the grocer for a reasonable price?

Will it be simple enough that I can press a button on my phone and have the steak ready by the time I arrive home from work?

Will I be able to order a pizza from a "designer" and pay for them to remotely print it to my house rather than deliver it?

I was not aware of that. This is an interesting counter-point to caf's observation: http://news.ycombinator.com/item?id=4188432

Yes, as I understand it (I'm no expert though), we can't distinguish between the background noise of the universe and an encrypted alien signal. I believe the main problem has to do with all the various possible sources of background noise constantly moving and interfering with each other through the galaxy.

That being said, from a physics perspective perhaps there's some way to say "these bits of random noise are unusually powerful, and based on our multiple noise detectors over a large area we've managed to estimate the path these signals travelled and there's only a planet on the other end with nothing else appearing in in the way (over the course of time it took for the signal to arrive) to explain them as mere background noise."

That's a good point. In effect, if we look at this from the perspective of the OSI model, you're saying that whereas the application layer may contain encrypted data the lower layers may contain structured data. So the logical question that follows: is there any reason to believe aliens might use encryption all the way down to the data link layer? Maybe our neighbours are all at war and need to disguise and protect signals from each other?

Have you read ithkuil's comment? http://news.ycombinator.com/item?id=4188160

Yes, I believe that images (encodings of electromagnetic radiation of particular wavelengths) and audio (encodings of mechanical waves of pressure) are universal concepts aliens are also likely to be familiar with, whereas understanding the information encoded within the transmission requires interpretation within a context we are highly unlikely to have here on Earth.

This is just my intuition and speculation but I would guess somebody could formalise some support in terms of Kolmogorov complexity (http://en.wikipedia.org/wiki/Kolmogorov_complexity): there are only so many ways to represent an image or audio, and over time an intelligent race will discover the least complex method and use it. Therefore, alien races of equivalent levels of technical advancement in a given field should be able to decode raw (uncompressed and unencrypted) transmissions of common types of information (images, audio).

Any additional complexity on top of this method will probably be quite structured, e.g. with television signals we have portions of the signal that are blank because during that time the CRT of old televisions would be retracing and not displaying data. (Actually, I believe these "dead" portions of the video signal are sometimes used to send additional structured data, e.g. Teletext, which would make the structure of the signal even more apparent.)

I suspect it'll be much easier to look at alien video transmissions and figure out what they're about than trying to listen do the same with audio transmissions because video would provide objective context (assuming it's not just a video of talking alien heads, or whatever the equivalent of a "head" for them would be).

With respect to encrypted transmissions, if the entire protocol is encrypted from start to finish, and you're using the right cipher* then no, you can't distinguish encrypted data from random noise. This is the principle behind TrueCrypt's hidden volumes (http://www.truecrypt.org/docs/?s=hidden-volume).

* What's the right cipher? So far no nobody has found a distinguisher for AES, and that's in common use, so that's probably good enough to make an alien race blind to us; i.e. we could be picking up signals from a race as sophisticated as c. 1998 humans (http://en.wikipedia.org/wiki/Advanced_Encryption_Standard) without realising it.

I don't think there are many different ways to encode meaningful information. We all live in the same universe and I see no reason to believe that intelligent alien life won't be intellectually quite similar to us. While tricky without context, decoding extraterrestrial signals will inevitably produce something of interest (images, audio) even if understanding e.g. the audio itself will be tricky.

That being said, while I suspect that meaningful, structured information will be, by definition, structured and therefore obvious to spot in a broadcast, the real problem is that these transmissions are likely to be (a) compressed and/or (b) encrypted.

While compression formats may reveal a few hints that there is some structured information present, encrypted data will be all but impossible to distinguish from random transmissions. We may be picking up transmissions from alien life but be unaware simply because the data is encrypted.

I have no idea what proportion of our broadcast signals are encrypted today, but I suspect it's a fairly significant and growing proportion (privacy and intellectual property concerns being the primary reasons).

It's an invalid argument because not all projects are alike. Facebook and Wikipedia are indeed very successful projects but they have a massive advantage over the types of projects I deal with in enterprise environments. For instance:

- technical environment: the Facebook/Wikipedia/generally-any-web-startup architects get to choose their technical environment 100%; in most corporates you need to integrate with legacy applications and databases in a specific environment, often enforced by IT policy, and this imposes significant constraints on which languages will be feasible options

- interoperability: for large, modular systems developed by disparate teams over decades with little or no standards or co-ordination, a statically typed language can be absolutely invaluable for ensuring interoperability and discovering usage (documentation is typically non-existent in these environments, as are tests, and the latter is seldom - often impossible - to retrofit); even for new systems, static typing can save a lot of time when developing and integrating new work across large teams, in my opinion

- resistance: business and technical knowledge lives in developers who are part of an ancient stack, e.g. Microsoft or Oracle or Borland; switching to PHP or Python or Ruby etc. means convincing these people to learn very different new technologies (good luck - won't happen), figuring out how to interface with legacy stuff using a language outside of the stack, introducing new tool chains, etc.j

- time & budget: even if the existing developers are keen to learn, it's tough convincing the project manager you need an extra month to go through the upskilling process and that you need to spend time researching interoperability options, and that inevitably you will run into future issues, etc.; the problem is most projects are run and budgeted for on a local cost benefit analysis and very rarely would the benefit of switching languages for one project be worth it (even though, in the big picture across multiple projects, it may very well be)

An observation: whenever this debate comes up on the HN, the proponents of modern, dynamic languages are invariably people working in closed environments where they have the freedom to set standards, do things right from the beginning (coding standards & reviews, tests, documentation) and the ability to make technology selections across the board. Often this freedom is not a characteristic of other environments, and consequently the language used in a one environment may be unsuitable in another.

You seem to be blinded by some serious Microsoft hatred. "Microsoft Randomised Brand Name(TM)"? "Office" is "random"? I suppose you think "Apple" or "Linux" or "Ubuntu" or "Sun" or whatever it is you are a fan of is less random? Another ridiculous statement.

Your friend told you you have a virus? Well if you know that means "Security" click on there, and, logically being unfamiliar with the 4 choices you pick the first one, "Security Home". Bang, right there in front of you:

    "Download Microsoft Security Essentials for free.  Help shield your computer from viruses, spyware, and other malware."

If you're too scared to click twice, or if the word "Security" is unfamiliar in the context of viruses, type "virus" in the search box on the home page. First result:

    "What is a Computer Virus | What Do Computer Viruses Do.  Have you ever wondered what is a computer virus or what do computer viruses do? Learn how are computer viruses spread and how to prevent them."

Second and third results discuss virus removal.

That's the only specific use case you mentioned and I had zero problems finding the correct solution. There are plenty of legitimate reasons to criticize Microsoft (the 3 hours I wasted yesterday trying to get ASP.NET MVC 4 working is one of them). But you didn't list one valid criticism with a specific use case. The rest of your post is a vague rant.

"It's just a vast array of disorganised links, or more precisely, a whole bunch of vast arrays of disorganised links, with no sign of any rational information architecture or even basic scannability."

What a ridiculous statement. Of course it's organised. Want to download something to extend a Microsoft product line? Click "Downloads" and choose the product line. Need Support? Click "Support". Want to buy a Microsoft product? Click "Store".

Which use cases exactly are you imagining are not supported by this "vast array of disorganised links"?

Although the fonts in Chrome render very poorly on my machine whereas in IE9 they look great...

I don't even understand how the parent expects people to know what is normal without posts like these to provide (admittedly anecdotal) data.

While I think credit cards have great benefits (see stevenwei for an excellent summary), getting the card does have a consequence in my experience: every card I've ever had has an annual maintenance fee (perhaps not the case in the USA?).

I've seen some of Dan's great photos before and it's awesome to be able to learn from him. Some of the advice he gives are things I've learned myself and explained to people before, and much of it is also new so I'm pretty excited to get through this properly.

One thing I would add, having been through this before myself and seeing my girlfriend going through it now as she takes up photography, is dealing with how people respond to your photos.

It can be disheartening when you take what you think is a great shot and nobody else (on Facebook, reddit/r/itookapicture, Flickr, etc.) seems to be interested. These feelings can be exaggerated when you see the photo that everybody is talking about is a simple cat photo or of some exotic location or event you're unlikely to find yourself in.

It's important not to allow yourself to feel discouraged.You have to just stick to your hobby for the original reason you picked up the camera in the first place and not to allow yourself to side-tracked by a lack of validation. It's important to seek out the opportunities leading to photos that you treasure and to remember that if others also enjoy your output that's a bonus, but not a healthy goal.

I'm not sure if Dan has experienced this (he may not have been concerned by what others think), but I believe it could be greatly beneficial to bring this mindset to the attention of readers so that if they find themselves feeling discouraged that they are aware of what's going on and how to deal with it constructively.

That article mentions Stuxnet once and draws no relationship between the killing and the virus.

Surely there is a massive selection bias in this approach?

I also came here to suggest a Markov chain generator using your data. I know nothing about music but it seems your analysis also needs to include information about timing (A is followed by B within X ms, C within Y ms, etc.) and some sort of clustering information that would allow choruses to be produced (e.g. identify clusters + cluster timing then information about the chords within the clusters).

Even better would be if you guys released your data and allowed other people to come up with creative ways of analyzing and using it ;) I actually have an idea that I think could work quite well and I'd love to build it but lack the data.

YouTube is of a different class of complexity as faced by enterprise developers integrating millions of lines of code across disparate systems.

Not sure what your criteria was for your list, but something I found interesting is this fact about the city I live in: "Despite the relatively dry climate, Johannesburg has over ten million trees,[22][dead link] and it is now the biggest man-made forest in the world" [1]

[1] http://en.wikipedia.org/wiki/Johannesburg

Ironically, I struggled to read this article on a mobile device (iPhone) because it hijacks the horizontal swiping which occurs while I scrolled and kept redirecting me to a different article.