Been awhile since I worked at AOL. One of the funnier moments for me was when I started working there, someone referred to the client (the one they shipped via CD) as WAOL (pronounced - Whale). IIRC, the executable was waol.exe.
Jokes aside, they had some pretty fierce technology on the backend that, had it seen the light of day of Open Source, would have been the precursor to Docker and Protocol Buffers. All this back in the 90's.
Man. The only purpose of all for me back in the day was that you could call them up and they'd ship you a stack of 5.25/3.5 floppies so you could "share" "the service" with your "friends". i.e. Punch a hole where the write protect window should be and yay, free, really cruddy, media!
An interesting thought, I'll bet that many companies if they had published papers or OSS for the technology they built would have advanced the state of the art today by five to ten years...
Actually a lot of the services that are now "big" existed already for a long time.
I still can't believe that back in 2001 we have been building something like a german Netflix.
Also my colleague who sat right next to me thought it would be a great a idea to have something like youtube for music in 2002 (Soundcloud) And he actually had a working service but he gave it up :)
I proposed a user-edited online encyclopedia in 1996, while working at the New Scientist. Everyone agreed it would never work, so the idea wasn't pursued.
Two years later, Everything launched, a year after that, H2G2, and two years after that, Wikipedia.
Didn't Netflix have what every other service was missing ? Netflix had a deal that they could stream whatever they had on DVDs for a fixed price. So they had boatloads of streamable content, which was the major stumbling block of everything else.
umm... not so much. Youtube was already a roaring success with user generated videos (which included lots of dvd rips, tv captures etc. etc.) before they started engaging with hollywood on a legal basis.
Did you do any work with the FDO language? The world of WAOL.exe was like pure ethanol poured onto the pilot light of curiosity when I was a younger programmer.
I never got the pleasure :-). I did see the manuals, which were a couple of binders full of pages, on a colleagues shelf once. I suppose that was the forefather of xul/xaml+scripting language, though I guess we're still trying to sort that out.
I think Facebook is well aware, that it could be over tomorrow, that is why they will survive. AOL tried to grow, I think Facebook acts more reasonable. They know the others can not get them in the social network space, no matter how hard they try, if they just secure their market share with acquisitions and slowly increase profitability. They do not try to win in another domain as the new Microsoft, Google does.
In the end Amazon will win. Most of the others make their money by advertising or by selling technology to companies, which make their money with ecommerce or advertising. Amazon owns every part of this stack, which actually makes money. The bigger the Amazon brand grows, the less money will be left for the others. At the moment, this is not visible, because the whole market grows. Amazon knows this and just reinvests and waits.
AWS and competitors are very capital intensive businesses. While the software might be easy-ish to replicate with a medium sized software team, the capital required to set up and then run dozens to hundreds of packed data centers is prohibitive for most competitors. The cloud market will turn into the chip market, where the capital cost of fabrication facilities is an enormous barrier of entry to the market. That is why the only real potential competitors in the space are from companies with lots of cash to burn and who can leverage the same "we need data centers anyways" model that Amazon does.
But it doesn't have to be dozens to hundreds of packed data centers. It just needs to be a lot of smaller operations, and a few mid tier like Linode, providing the same service at various prices and scales that make sense to various sites.
AWS as a service is providing a great deal many more services at a much faster rate than much of their competition, they basically own public cloud. They were also first to jump the gun with stuff like ECS and are doing some interesting things around RDS with Aurora. I feel AWS will only become even more dominant.
This is not to say there are some advantages in a few others (say, Rackspace appears to be simple to grasp, excellent human support) and GCE (load balancers don't need to be pre-warmed, etc).
But if anything, I see it increasingly harder for the other providers to keep parity with Amazon, because of their velocity and market share.
Now, for many people using cloud as just a easy way to get IP addresses, it's all the same, but once you start using a lot more of their services, they provide a lot of good stuff in what is basically "middleware" for applications to latch onto.
The question is whether an organization is doing this themselves (i.e. run their own Foo on their nodes) or is using the value added services or not.
EC2 and S3 are't trivial to replicate. Moreover, consider the entire platform and the dominant position of Amazon. Amazon is the Microsoft of yore- they actively cannibalize any profitable service on their platform. How many products is AWS, now? Perhaps 50, or 100? Can anyone reasonably migrate off AWS after integrating with even 5 or 6 of them? Did the profits of Office go to zero as it was copied and handed out for free a dozen times over? Amazon will be making bundles of money with AWS for a long time to come.
For all intents and purposes, Microsoft Azure and Google Cloud Platform have already replicated EC2 and S3, and every other feature they think is worth it. The price war in 2014 was pretty brutal (unless you're a customer, of course!) and it's not going to slow down in 2015.
And as people move away from 'raw cloud metal' to whatever the layer du jour is, what's underneath will matter less and less.
Lots of people concentrate on the Fire Phone, which flopped. Check out other recent Amazon efforts - Fire tablet, Fire stick, Amazon Echo. The first two are solid competitors in their respective spaces (ereader/tablet, streaming media), while the Echo is a pretty amazing effort for a hardware category that didn't exist one year ago.
I have an Echo. Showed it to my kids. They said, "It's like Siri but she can understand us!"
Of course, they were upset that she wouldn't respond to the funny sort of queries you could make of Siri. And my son really wishes her ability to do conversions extended to questions like "how many miles are in a light second" (Siri uses Wolfram Alpha to great effect here. Alexa - Amazon Echo - just apologizes and says she's added a bing search to the echo app.)
They earn money, they just choose to spend it again to earn more of it. This is entirely different to most internet businesses which get money and try to earn it by spending it.
I know relating things is very natural, but I don't feel Facebook is the new AOL at all until they offer ISP and a web browser. That is what made AOL AOL. Facebook may be many peoples main destination, but it isn't how they connect to the internet, it isn't primarily a gateway to other sites. Facebook doesn't want you to leave Facebook at all.
> Facebook doesn't want you to leave Facebook at all.
AOL didn't use to let you leave AOL. For many years it was a 100% isolated walled garden. That is what made AOL AOL many years before AOL became an ISP.
They only grudgingly started adding ways of accessing internet content as the growth of the net so outpaced AOL that they had no alternative in order to keep subscribers.
And, if memory serves, when they first began providing access to the Internet they also charged extra for it (although I may very well be mixing them up with someone else).
I think the analogy is about being the sole portal and walled garden, rather than a browser. For a long time AOL didn't have the browser, just internal content. It morphed from online for the leaders to internet for grandma and grandpa. Going too much beyond this will break the analogy.
Facebook just achieves lock-in in a different way, AOL forced its subscribers to use their portal and aggressively tried to hook people by providing free samples. Facebook has captured enough of the market that people "voluntarily" only interact with their online friends through facebook, or rather they are forced to due to network effects and spend a significant amount of time consuming "content" and ads.
It also turns the dynamic of the web, where you actively seek out information through search and by following links, on the head, by pushing information to you, forcing you to at least acknowledge them. It has a similar ad-ridden, homogenous aestetic as AOL, practically unbearable without some ad blocker.
Facebook is the new Myspace. Same basic concept. Same growth in ad density over time. Same decline in customer experience.
AOL is still a dial-up ISP, with 2 million customers still paying $20.48 a month. That's where their profit comes from. Their new junk-content business turned out not to generate much profit. (http://www.washingtonpost.com/blogs/the-switch/wp/2014/08/07...)
In one of the bargain-binned dot com frenzy era books, I remember an AOL executive dismissively claiming that they (AOL) could put a AOL login or logout popup ad that said "Click here to pay $19.99 for absolutely nothing" and some fraction of a fraction of their customers would buy.
It cannot be ruled out that those kinds of ads won't ever come to Facebook. Ads that buy with one-click (despite Amazon's patent) and rely on partial user confusion and deception as a business model.
If Facebook did that, or even if advertisers did that and Facebook didn't kick them off, it will put the brand in serious trouble, they'll get fined, etc.
This is the sort of thing small scummy businesses do where the owner can hide behind it, go bankrupt and start again. It's not something a top company would do.
I am not saying big companies won't do evil things. This isn't going to be one of them.
And also you don't need to confuse people into paying for nothing. You can do it with their eyes wide open. In-game purchases, for example.
In regards to Qualcomm dominance, thankfully there is no total domination yet - especially in the lower/midrange segment, where there is MediaTek (heh, and MediaTek is the 4th largest IC designer worldwide).
And actually, I prefer MediaTek phones - you can root every single one of them and in contrast to other phone manufacturers, their preloader (first stage bootloader) even allows reading back the entire ROM. No restrictions. (For the interested, google for spFlashTool)
My 3 last phones have been MTK. The price/performance is much better in my opinion than most of the alternatives. Of course I won't get top range phones, but I'd rather get a new MTK based phone on a more rapid update schedule, and recently there's been some very high quality phones coming out (e.g. aluminium casing, gorilla glass and very high quality full hd displays) using MTK chips.
My one caveat with them have been the dreaded weak GPS, but I don't need GPS much, and with my last phone that works fine too anyway (but anyone considering MTK based phones: if you need GPS, read reviews; on some it works fine, on others it requires some light modding, and on some it's practically impossible to get the GPS working without sacrifices to elder gods and lots of swearing).
Mediatek even seems to supply the entire software stack for the phones - and the hardware design part... well, just google for "gongkai" (http://www.bunniestudios.com/blog/?page_id=3107). No surprise that the speed of innovation with Mediatek chipsets is far bigger than the behemoths (Samsung)
I agree. Qualcomm is dominant, but not quite "Intel-dominant" in mobile, especially when Apple makes another few hundred millions chips of its own.
That said, it's becoming increasingly worrisome that so many OEMs just "default to Qualcomm" these days. If I were Samsung's CEO I'd put much stronger focus on making my own chips and putting them in my own devices (just like Apple), while also selling them to others, just like they sell their Super AMOLED displays and so on. I would also buy out AMD, mostly for their GPU expertise, but also to enter in the server/data-center market, which is more profitable than the mobile market. Samsung has the money and marketing clout to do much bigger things than AMD can do right now alone.
I don't know what the hell Nvidia is doing in mobile, but they always seem to screw something up. Their non-focus on energy efficiency in "mobile" and their focus on "performance" for umm..tablets, has proven to be a mistake with every single chip generation. Stop it, Nvidia! You need to win smartphones, not tablets, which are a dying breed already. And to do that you need low-end chips as well not just "PC-class mobile chips", otherwise you'll never get the market share you need to help you become "popular" in the mobile market.
And finally Intel - oh, Intel, how I despise thee. Two main reasons why I don't want Intel to win even 5 percent of the mobile market. Short version: 1) they are anti-competitive, 2) they don't deserve it.
Long version - Intel has started having quite the "monopolist culture" a while ago (at least a decade). They've learned that playing the monopolist card when in power "works" (to crush competitors and entrench themselves in the market), and they try to use it as often as they can and to the limit of the law (or even well beyond it, if no one is paying attention). Intel has almost wiped out GPU vendors from laptops through strongarming of OEMs and shady pricing tactics and, with Broadwell and Skylake having an even bigger focus on GPU performance, I predict within 2 years no one will even hear about an Nvidia GPU in an Intel-based laptop (unless we're talking $2000+ machines - maybe).
They also have a monopoly in CPUs for PCs and are now using the profits they get there to sell their their Atom chips below cost (a practice that's illegal in many countries, but either they aren't paying attention or Intel manages to twist that into some kind of "business deal" they are making with OEMs).
That's just the anticompetitive part. The "don't deserve it" part comes from the fact that Atom sucks compared to the highest-end ARM chips, even though it's built on 22nm FinFET, and those ARM chips are built on 28nm planar (if we're to compare them at the time of market launch) - essentially a 3 year/one node and a half process advantage for Intel. So they are trying to push bad products into the market and gain market share by selling them below cost to trick some OEMs into accepting their chips. And I say "trick" because that's what it is - at best, a short term advantage of them, because if Intel would ever have even 20 percent of the mobile chip market, those same OEMs would have to pay them more than what they pay for ARM chips, because it's "Intel".
So many other shady Intel practices such as tricking consumers into thinking "Celeron" chips are Celeron, when in fact they are Atom, lying about the real TDP/power values of their chips to make it seem like their PC chips are almost comparable to ARM (while being misleading about the fact that they also have much lower performance) and so on. Intel is a shady company through and through when it comes to selling its products.
> Intel has almost wiped out GPU vendors from laptops through strongarming of OEMs and shady pricing tactics
You forgot one point: good (open-source) drivers. Intel drivers Just Work (tm), no matter if the OS is Windows, Linux or OSX. AMD/ATI drivers... not so much, and NVidia's work, but are still closed-source.
And the "The drivers Just Work" doesn't end for Intel in GPU drivers, but also the other driver classes - just take wireless and chipsets. One driver package for all, supplied by Intel, no need to hunt down obscure czech or russian sites for drivers (hello, Atheros).
You shouldn't need to look very hard for Atheros drivers, since they're often included in the kernel. At the very least, they're in your distribution's repository, since they're not proprietary.
I'm talking about Windows here, to be honest. I use both Windows and Linux and, well, when I configure laptops, I always use Intel (and NV for the GPU) components because they are guaranteed to work without spending hours on Google.
Ooh it really grinds my gears imagining how much Intel might charge for mobile chips if they hypothetically had more market share. Oh that imaginary Intel really will stop at nothing!
Like the way they illegally sell their Atom chips cheaply and get away with it! They probably do it legally somehow but don't let that stop you hating them! Grr! Grr!
Come on, look how the competitive Atom chips are made with better technology! Why can't Intel use worse technology like Arm does and then we'd see how ... worse their chips are after forcing them to be worse!
Interesting, and fair point. I tend to type things explicitly for the opposite reason, because I trust my fingers more than some obfuscated typo in an URL - but I see logic on both sides.
OTOH normally when I ask the people I see doing this why, their answer is more along the lines of "dunno" :)
I think we are hitting the point where the technology is there, even if it is a bit rough. Unfortunately it is still bleeding edge, and it seems like adoption will take a while.
I agree. We may not even be seeing the Rift CV1 until much later this year. Who knows what the experience on that will be like - DK2 is close but not quite there.
Jokes aside, they had some pretty fierce technology on the backend that, had it seen the light of day of Open Source, would have been the precursor to Docker and Protocol Buffers. All this back in the 90's.