Jacques Mattheij

technology, coding and business

The Pendulum Swings, Again

In the early days of computing, computers were physical monsters, occupying large rooms or even whole buildings. Acess to them was closely guarded, you were lucky if you had access at all, let alone close-up and personal access. That was reserved for the system operators, the people tasked with the relatively boring job of babysitting the machinery, fixing it when it broke and shunting jobs in and out of the queues.

The name of the game at the time was batch processing, and it made the people tending the computer look somewhat like a priesthood. Access was valuable enough that computing jobs were accounted by the second and if your job used more cycles than you were alotted then tough luck. The age of the mainframe was long and if you were interested in computers but didn’t have a leg up in some giant company you were essentially relegated to programming on pieces of paper and running your code by hand (a useful, if somewhat tedious skill).

Then several revolutions happened in relatively quick succession. The first was that as computers miniaturized due to the increasing density of the logic circuitry that powers them that suddenly departments found that instead of going to the high priests they could instead afford to buy and house a computer of their own. This was the - brief - age of the mini computer. Examples of the devices sold were the VAX series (by Digital Equipment) and the Nova (by Data General).

Such smaller computers did little to improve access by ordinary mortals, but they did give rise to a much larger number of people dreaming of such access, and they caused a further spreading of the knowledge required to use a computer. Sooner or later the inevitable happened, hobbyists acquired the skills required to create a simple computer with only limited processing power and a limited amount of RAM, but they were all theirs to play with. Thus the era of the personal computer arrived.

There was no DRM (people were happy that stuff worked at all, frustrating someones use of a piece of data or program seemed completely weird, it certainly still does to me). Knowledge was freely shared between those that had it and those that didn’t and soldering irons and computing still went hand-in-hand. If a computer broke, you pulled out the schematic (typically on the last sheet of the manual in a generous display of full disclosure and not a little pride). Apples, pets, atoms, bbc’s, commodores, TRS 80’s (and numerous clones such as the UK dragon), Timex-Sinclair and on and on. It was a veritable cambrian explosion of computing gear and associated peripherals. Assembly ruled and life was good.

I think this was the point at which computing was at its most democratic.

The first reversal of the movement of the pendulum happened roughly about here.

Most of those 8 bit machines are long forgotten - except by old people like me ;) - but they missed something that those big computers from the old days didn’t. They didn’t come with someone to blame when stuff didn’t work, and as such they were unusable for most businesses unless the people working there wanted to become intricately familiar with the hardware and software development themselves. Most businesses saw computers as a neccesary evil and it took the force of none other than IBM to formalize the personal computing world to the point where businesses decided to join the fray.

The PC was a runaway success. And so were the clones! Olivetti, Compaq, HP, Dell, Apricot (original name that one) and many many others. Untold number of them sold, and they in turn paved the way for more and more infrastructure and computer literacy. To the point where there was a PC in almost every home in the developed world. Those PCs were used for everything from running a small business to entertainment in the form of computer games or balancing the checkbook. Limited forms of connectivity were available, mostly in the form of BBS systems using dial-up services at glorious speeds of 300 or 1200 baud (technically, not equal to bits-per-second since a single baud can encode more than one bit but we can simpify this to ‘terribly slow’ for those not comfortable with data comm terminology). AOL and some other companies like it created a walled garden within which the people using it could communicate with each other.

Commercial software was the norm, even if there were experiments with things such as shareware and freeware. On the whole, if you wanted a computer program, you wrote it yourself, you paid someone to write it for you or you bought it. Many a fortune was made in the commercial software realm. Some names from that era: Microsoft, Borland, Ashton-Tate, Paradox, Clipper, Norton and so on.

Then, in 1993 something quite amazing happened and it caused the pendulum to revsere direction once more.

Up until that point access to the internet was limited to people working for specific government entities, universities or affilated groups. Regular joes such as me - and probbaly you - didn’t have a chance at connecting to the internet. But in 1993, for the first time regular people could subscribe to a service which gave them dial-up access to the internet. That, combined with the NCSA Mosaic web browser made it possible for ordinary people to join in the online information revolution.

Not all those people were active on the creative side of things, but enough of them were that the number of web-pages and hence the amount of information available at a moments notice to ordinary people increased very rapidly indeed. In turn this engendered a new wave of democracy in the realm of software and information.

The free software movement (founded much earlier in 1985) suddenly found itself with the ability to harness the power of the tens of thousands of computer literate ‘loners’ out there and free software of amazing capability and complexity was churned out in very large volumes.

In 1995 this swiing of the pendulum was further accelerated by the lifting of the restriction that the internet should not be carrying commercial traffic.

Media files, text files and programs. It was all bits and bits can be transported halfway across the globe, copied without any cost related to how much money went into creating the bits in the first place.

Big business took notice of this new phenomenon called the world wide web, and the pendulum swung again.

This time it swung hard, large amounts of money were at stake and the internet/world-wide-web combo threatened to upset more than just a few very valuable apple carts. Nicholas Negroponte’s book ‘being digital’ woke up a large number of slumbering giants to the fact that their future may not be as secure as they once thought it was. And the harsh reality of some early cases of disrupture proved it beyond any doubt. Governments were getting worried about the power of the internet and many laws were passed in order to try to bring the situation back under - their - control. Big media didn’t know whether to love it or hate it and a period of confusion reigned. But little by little the amazing freedoms of the late 90’s gave way to various forms of oversight, as well as wire-tapping and legal attacks on citizens at unprecedented scales.

Computer programs would no longer run on hardware that they were in principle compatible with without certain keys to unlock them (a similar thing using hardware had happened in the 80’s with so called ‘dongles’, typically for expensive programs such as CAD software).

Programmers that wanted to write code for some platform or other had to have their software blessed by gatekeepers or they’d be totally out of luck. Those very same gatekeepers wanted a piece of the pie for the right to access the users of their devices and wanted a cut of media sold for consumption on their devices regardless of whether or not they had a hand in the creation of those media. The line between ‘computer’, ‘program’ and ‘media’ blurred to the point where consumers were successfully confused about what it was they were actually buying.

And that’s where we are today. I think that the ‘openmoko’ phone was an (too) early attempt at forcing a reversal. The raspberry pi may be a key component here (it shares a lot of its heritage with a large number of smarthphones). The future of computing lies with mobile devices, and I sincerely hope that it lies with open mobile devices rather than the hodgepodge of closed and half-closed systems that we have today.

As you can see from the list above every swing of the pendulum was more violent and abrubt than the one preceding it. This is in part due the larger number of players involved and due to the larger commercial interests involved. But I can’t help having the notion that we’re heading for some kind of escape velocity, after which the pendulum will swing through so hard that it will get lodged to remain in that position forever after, or at least for a very long time.

If my observations are correct then such a swing is about to happen, and this time we had better get it right. Things that point in the direction of a swing are an increasing awareness of ordinary computer users with respect to their privacy and who actually owns all that data. The fragmenting of the smartphone and tablet markets will lead to some more openness and at some point all the bits and pieces to create true open hardware will fall into place. FPGAs, 3D printing and some more ingredients will make it possible to generate very capable hardware and versatile on a relatively small budget. A mobile platform consisting of a standard phone/3G/wifi module, an FPGA and a standard touch screen would be pretty formidable and is almost within reach. These and many other bits and pieces hint at yet another round of change. It may very well be our last chance.

Remember that there are two possible outcomes, one where the internet successfully manages to cause a swing to the edge of freedom, and another where it is successfully co-opted by big money and governments in a concerted effort to give us all a subscription to online Life-As-A-Service where you will be beholden to some party for the ability to gain access to knowledge, information, the right to communicate and so on and where the act of programming will be as tightly regulated as the export of cryptography was.

Those who lived through that period and paid attention are probably well aware of the ridiculous notion that information flows should be controlled in order to protect the children or whatever the fig-leaf of the day is. Ask the Iranian and the Chinese people how they like their firewalls.

The bit that worries me is that the tools for such wide-spread decentralised and automatic control weren’t available in the past. But they are becoming available as we speak and the outcome of this current move of the pendulum is anything but sure. Let’s make sure it ends up on the side of freedom and democracy, and not on the side of DRM and other forms of control and digital oppression.

Edit:

Astute readers at Hacker News pointed out that I’d missed the opportunity to list those things that make me believe that another swing of the pendulum is about to happen. So, here is a list of items that I came across in the reasonably recent past that give me some - but not a whole lot, admittedly - hope that we may have one more shot at getting this right:

  • http://en.wikipedia.org/wiki/OpenBTS An open source 3G base station (the spectrum is of course still regulated but there may be ways around that)

  • The 3D printing revolution, such as embodied by the reprap: http://www.reprap.org/ This could do for physical objects what the fax did for paper. Given a 3d scanner and one of these gizmos you can copy any device of which you already have an instance, or you can download a 3D model and print out something someone else has (or has designed) on the other side of the globe. These little machines work on an additive process so they do not leave you with a mess of swarf to deal with but just the finished object.

  • FPGAs. I believe that as prices of FPGAs will come down that they will open up a completely new frontier wrt to the development of new hardware. Basically FPGAs are a way to reprogram a chunk of hardware to become a completely different device. Just like in the eproms of old you re-wire the circuitry, but instead of just giving you data reprogramming an FPGA changes the way in which logic cells are wired together. This is a major step because it takes away the necessity of having a functioning fab or access to one in order to make your own chips and that in turn is a huge step forward for open hardware.

  • Consumer dissatisfaction with the state of privacy on the internet. Even my mom complains about facebook and google spying on her and believe me, she’s got ‘nothing to hide’ (at least, I think so. Hm…)

  • https is on the rise, more and more parties are switching to encryption by default. Strong encryption in the hands of private individuals has already been classed as a munition once, I’m beginning to see the wisdom of that classification. Just imagine, all those private citizens communicating without the ability to listen in on the conversations. Of course there is still traffic analysis to be performed and who talks to who is a very important chunk of an investigation but that’s a step up from having the plaintext in front of you. At some point I think that encrypted email will be the norm. Now if everybody would just stop using gmail the privacy that you traditionally could expect from a letter may even return.

  • The raspberry pi. As another commenter on HN pointed out, the raspberry pi is only ‘half open’, I missed the fact that not all the software is open, I knew that the specs for the broadcomm chip were not available. That said, there are two things about the raspberry pi that I greatly appreciate: the price point is about as democratic as it gets, and the people behind the project are imho absolutely committed to doing the right thing if they have half a chance. Eben, Liz and the other people behind the project have already shown that they will do whatever is in their power to open up the design as far as they possibly can without violating their contractual arrangements with broadcomm. Maybe one day they’ll get broadcomm to open up their design, maybe one day they’ll produce a ‘pi’ based on a new chip or an FPGA of their own design. Economies of scale favor the broadcomm route at the moment, clearly it packs a lot of punch for very little money and if that enables a veritable legion of new programmers to gain access to enough kit to do some serious damage then I’m fine with it.

  • Mesh networks are on the rise. Of course, hardware and software by themselves are not enough. But the wifi spectrum will be hard to take back and with every router that appears there is more potential to string them all together. A hackable router (and there are several of those) could easily become a node in a mesh network. There is a lot of research to be done in this field before mesh networks can become a practical reality at acceptable speeds but the beginnngs are definitely there. Looking forward to the first attempt to restrict the deployment of mesh networks in a legislative body near you.

  • Flash storage is coming down in price to the point where something that fits under your thumbnail can contain a body of data that a librarian a few years ago could only have dreamt of. The whole production of the move industry for a couple of years now fits comfortably on a device that weighs less than 10 grams and is nearly indestructible. As the value of the storage carrier drops the ease with which large volumes of data can be passed out to leak it or make it nearly impossible to censor it will increase. Smuggling a harddisk out of some installation is tricky, smuggling a micro SD card out is childs play. The penalties won’t be any different but the chance of detection is vastly smaller.

  • Academia seems to wake up to the idiocy that is the textbook racket and the distribution of scientific papers. More and more teachers and scientists distribute their materials for free and publish in places that are not under the control of the likes of Reed-Elsevier.

  • The use of DRM on media is dropping, it looks like that battle at least was won. For now. But there is enough money behind the media industry that I expect periodic revivals of attempts to re-introduce it through a legislative move instead of a simply technological one. Remember when Reverse Engineering was legal? It SHOULD be legal, but right now it isn’t. In part that stems from the attempts by the media industry to stop you from reverse engineering copy protection. Then there is the ease with which media entities can now routinely claim copyright infringement even when that could only point to specific instances of data that was copied. But somehow this now applies to torrent files (which do not contain any copyrighted content themselves) and to such things as simple hashes (magnet links for torrents, for instance, but this has also been used to attempt to limit access to keys for decrypting media files). That’s insanity, and we surely have not seen the last of this. But for now at least, the DRM battle seems to have been won by the good guys, I’m not betting the farm on who will win the war.

  • And many more… I’m optimistic by nature so I tend to think my glass if half full by default but there is a serious undertone here. Yes, we have all these goodies and yes, we could extend this list until it has 100’s or even thousands of entries. But the volume of entries in this list will not eventually make the difference between which side will win this. That will be decided by whether or not the masses can be educated to the point where they will choose what’s right, and not just what is a momentary convenience. The internet is a very powerful tool. Like every other tool, you can use it for good purposes and you can use it for bad ones. Right now we’ve handed way too much power to people whose intentions are commercial at best and downright evil at worst, even if they’ll claim openly that that is absolutely not the case. AOL or Facebook, Google or Apple, Microsoft or Oracle. I find it hard to the see the difference between any of these parties and the political parties that drive the deals that allow them to spy on the citizens of the planet and to abuse their data for net negative purposes.

There is no sense of honour in this any longer, it used to be that you lived by the letter of the law, especially as an elected official or a responsible business person. That has changed. Nowadays living by the letter of the law seems to be interpreted as ‘as long as you can get away with it and you don’t get caught’. If we don’t want to end up under some collective boot then we need to wake up to the fact that all these shiny goodies and proprietary stacks containing our personal data will likely be used against us and that we probably have another shot at reclaiming what is ours but that there are no guarantees and that there will be a lot of headwind.

Open Source software, Open hardware, open networks. Anything less simply will not do.

HN thread on this article: http://news.ycombinator.com/item?id=4563913

End Edit.

A list of urls that may be of interest if you thought this article was worth your time:

http://en.wikipedia.org/wiki/Trusted_computing

http://en.wikipedia.org/wiki/Free_software_movement

http://en.wikipedia.org/wiki/Phil_Zimmermann

http://www.fourmilab.ch/documents/digital-imprimatur/