Archive for the ‘Egypt’ Category

It says something about the modern thinkers that one of the Egyptian spooks the Piggipedia team identified turns out to be working in the “Security and Loss Prevention” department of a major hypermarket chain. Having lifted photos with names on them from their HQ, they started searching PofacedBook – sorry, LinkedIn – for them, and discovered many more stories like this. That particular spy had been in charge of infiltrating NGOs before shifting over to the retail sector. Sadly, they started deleting profiles pretty quickly.

Here’s Yahoo!, heroically defending the image rights of individual Egyptian torturers. It looks like it’s getting time to slurp all my stuff out of Y!, but beyond that, what a thread. The number of people queuing up to explain how compassionate and reasonable the boss is being. Ugh. Frankly, I think you might as well post the damn photos if you stormed the secret police headquarters.

So you may as well support the Piggipedia and write to them at this e-mail address: case982056@support.flickr.com

PS, after all the cock about Facebook revolutions, the US State Department’s taking its sweet time to call off Yahoo!’s intellectual-property Brian Coats. You might think they were embarrassed about potential disclosures.

This post brings several things to mind. Apparently, eastern Libya was a hugely overrepresented area among the international jihadis who went to Iraq and there exploded. Clearly, this means that you can’t assume that they’re fighting for democracy, whiskey, sexy.

However, it’s also very likely that this represented a deliberate policy on the part of the Libyan government to channel its dissidents into particular ideologies that its new friends also perceived as the enemy, and then to ship them out of the country and hope they would explode somewhere else. Making jihadis – repressing all other forms of dissidence, while not trying too hard to stop them recruiting or leaving the country – had the side benefit that it validated their claim to be a bastion of stability assailed by Islamic extremism. They could produce the extremists, after all. And it further allowed them to avoid burning all their bridges with the other side. If it became expedient to make friends with the terrorists again, they could produce the bloody shirts – the martyrdom videos – and demonstrate that they had been useful.

Of course, Gadhafi didn’t have to be an evil genius to come up with this plan – he was essentially copying Saudi Arabia’s homework, and depending on how you look at the relationship between the Egyptian regime and the Brothers, perhaps sneaking a look at the neighbours’ as well. Giddens may have thought they were going to be a new Norway, but the real plan was more like Saudi 2.0, probably right down to the hereditary government.

Another lesson from this is that they’re probably not going to give up easily.

Sultan al-Qassemi kicks in a data point to the ArseDex. Apparently Libyan agents are distributing flyers in Guinea and Nigeria calling for mercenaries to fight for $2,000 a day. Yesterday, loyalist thugs cost $500 a day in Libya. Even with the huge supply of potential thugs in sub-Saharan Africa’s demobilised militias being available, the ArseDex has gone non-linear – it’s risen by a factor of four in 24 hours. Arseholes now command a premium of four hundred times the average wage. Surely Gadhafi must be doomed now.

The data’s pretty sparse, but here’s a spreadsheet. The edit link is here.

https://spreadsheets.google.com/pub?hl=en_GB&hl=en_GB&key=0AjP2Zn6KkPUwdFpCd1ZITlpMU3FLYk91RExxaDZvUUE&output=html&widget=true

Pulling together various resources, I’m beginning to get a picture of what happened with the cut-off and restoration of the Internet in Egypt. First up, at least in some senses, it may be valid to say that the Internet played a role – Arbor Networks observed that traffic to and from Egyptian networks (and between them, in so far as any of them are customers of Arbor’s) had spiked dramatically, almost vertically, in the two hours before the cut-off and that the whole week up to the 28th of January had been one of unusually heavy traffic.

When the cut-off went into effect, at 5.20pm local time on the 27th, it was implemented by forcing all the networks that peer at the Telecom Egypt-controlled Internet exchange to drop their BGP peering sessions with the exception of AS20928, Noor Data Networks. Famously, this is the operator that serves the Central Bank and its payments settlement system. Essentially immediately, 2,576 networks announced by 26 Autonomous Systems became unreachable. The surviving 26 ASNs including, as well as the Central Bank, the Alexandria Library, and the national research & education network, which if it is at all like most NRENs has a lot of its own infrastructure.

On the 31st of January, there was a further wave of cut-offs which removed another 14 ASNs and 134 networks. The list of the last survivors is here – notably, someone had clearly realised that not cutting off the students, of all people, was a missed opportunity, as the NREN isn’t in there. However, one of the mobile operators (UAE incumbent Etisalat’s national opco) stayed online although they had been ordered to cut off the mobile service itself. Perhaps they provide service to the government’s mobile devices?

Interestingly, however, according to posts to NANOG, several of the .eg root DNS servers remained online (not surprisingly, as at least one is outside Egypt). Even more interestingly, even after the BGP sessions with the IX were pulled down, the lower layer equipment stayed active – Egyptian ISPs noticed that there was still link light on the fibre optic lines between their locations, and theoretically it would have been possible to cobble together static routing between their systems.

Similarly, the internal voice network remained operational and so did the international SS7 gateways that link it to other phone systems. As a result, some people found that they could still reach their ISP, whether by dial-up over the voice circuit or even sometimes on DSL. The question, though, was whether there were any routes beyond the ISP’s nearest point of presence. Several foreign ISPs offered free dial-up connectivity over international phone service (notably this French one).

And, it seems, Egyptian ISPs also tried to re-establish internal connectivity after the cut-off, when they noticed that the fibres were still lit up. However, the problem was more subtle than just pointing static routes at each other. Communicating with people outside Egypt wasn’t, after all, the primary need, and anyway, it required passing through the government-controlled exchange.

But the problem with Facebook, Twitter, Gmail or what have you is that unless they have data centres in your country, they’re international traffic. Depending on their internal architecture, even if they do, they might be dependent on international routes. An Egyptian engineer who posted to NANOG during the revolution made the interesting point that, although Egyptian ISPs are relatively well-interconnected among themselves, not that much traffic flows over the interdomain links as so much stuff goes out to the global Internet. It’s analogous to the old problem that the topological centre of the African Internet was 36 Tooley Street, London SE1 (the LINX headquarters), or 111 8th Avenue, New York, depending on whose version of the story you like better, although less pernicious as the infrastructure is there to solve it.

Sometimes this is useful – it’s harder to censor stuff hosted in another jurisdiction. But it’s also a problematic dependency. Back in the Egyptian NOC the New York Times was hosted on, they were struggling to find copies of key software packages to distribute, for example clients for Internet Relay Chat messaging, and also critical data files such as cached DNS zones, lists of domain names and their corresponding addresses. Some ISP engineers are now working on preparing emergency packages of software and data for use in an extreme emergency – for example, regular dumps of the root and local DNS zones, similar snapshots of the local routing table, not to mention PGP signing keys and contacts for as many other engineers as possible.

After all this, what were the government’s aims? The initial cut-off was probably motivated by a combination of wanting to black out sources of independent information and hoping that it would hinder the protestors’ organising. Some of its particular details – for example, leaving 20928 up and not trying to shut down interdomain links within Egypt – may have been an effort to keep some “normal service” going, as well as not preventing VIPs from transferring their money out of the country. It’s also possible that cutting off link light between all Egyptian ISPs without physically grubbing up the fibres was harder than it looked.

So then, why did they bring it back on the Tuesday of camels and thugs? One interpretation is that they were hoping people would go home and update their Facebook statuses, which would have been incredibly patronising. But the Egyptian elite patronised the hell out of the public every time it went on TV, so it can’t be ruled out. Another one is that they hoped to project an impression of returning normality, which didn’t really fit with thugs on horseback swinging knives, but then their response wasn’t characterised by coherence.

Another still is that they hoped it would help to get the government’s propaganda out there. This argument – Gamal Mubarak flipping through his copy of The Net Delusion in a curtained backroom of the palace – has the advantage that when the Internet and the mobile networks were reactivated, there was a rash of reports of loyalist trolls, and one of the first things that happened was that the government forced the mobile operators to send out threatening bulk SMS messages – spam as a weapon. But this was surely incredibly optimistic.

In fact, what did happen was that people started doing precisely what they had only been doing to a limited extent the week before. Twitter feeds from Egypt filled up with what the NANOG crew would term operational content – requests for more medical supplies, reports of a lost child, calls for more protestors to mass at a specific gate into Tahrir Square. This was the real thing – a tactical radio network for the mob – and ironically it was mostly running over SMS and going out to servers elsewhere in the world. And, of course, its major carrier was the much reviled Vodafone Egypt, unwilling deliverer of Central Security’s spam blitz.

Thanks to reader Koranteng for this data point. You may recall this post about the market for thugs in Egyptian politics. Specifically, when the government needed arseholes to attack the protestors, it had to pay four times per capita GDP to get them.

In Libya this week, it is said that the government is using mercenaries recruited from its various allies’ wars in sub-Saharan Africa as arseholes, and that it’s paying $500 a day for their services. Libyan per capita GDP is $14,884 at purchasing-power parity, so the price of privatised violence is running at a premium of over one hundred times typical earnings. Clearly, either the regime has so much less real legitimacy, or the degree of brutality required and risk involved is that much higher. In fact, those options are both consistent, as a regime with less legitimacy would need to use more force and it does seem to be doing just that.

I made the point last time out that it’s typical for mercenaries to be very highly paid relative to the countries in which they operate. This is clearly an important point here. It’s also true that Gadhafi’s Libya has often got other people to fight its battles for it – they exported Palestinians into a variety of different wars in the 1970s and 80s, notably sending PLO volunteers to prop up Idi Amin (you bet they didn’t sign on for that). Later, in the 1990s, they trained and equipped fighters in the various West African civil wars (notably Charles Taylor – there’s an arsehole for you). Now they’re doing the opposite.

Of course, being an oil state, they can probably afford to keep hiring the arseholes.

However, here’s something interesting from the Egyptian elections last year, from Reuters.

Rates for hiring a thug start at 800 Egyptian pounds ($140) and can reach 40,000 pounds depending on the assignment, according to a study printed by the independent Wafd newspaper.The study, by criminologist Refaat Abdel Hamid, said thugs hired to attack large groups or candidates cost 25,000 pounds a day. Those hired to resist the authorities cost 6,000.

“The price of thugs includes compensation for custody and hospitalisation,” the study said. “Former and current ministers and the NDP party get special prices and discounts. Prices are hiked for businessmen and first-time candidates.”

That suggests that in October 2010, your entry-level goon came in at about twice the rate Mubarak was paying at the height of the revolution. Interestingly, if you were looking for goons who would be willing to assault a crowd of rivals – the same mission the camel riders had – you’d have had to pay much, much more. Thirty times more, or perhaps there’s a zero missing somewhere, in which case it would imply an even bigger price drop. Part of the difference might be explained by the NDP claiming mates’ rates as a large customer of long standing, and one who could offer valuable side payments in the event of success.

But it’s hard to think of any explanation why the NDP would have been paying less for thugs at the height of the revolution, when they would presumably have been in demand, and the party itself would have been desperate. Also, assuming the selling party could read the writing on the wall, they would surely have been likely to insist on payment in cash on the nail, rather than promises of future side-deals that would likely never be fulfilled. Perhaps the supply of potential thugs increased, but how? Was violence just a more salient possibility?

Or perhaps there was a radical shift in the supply curve between October and January. If the usual sources of goons were for some reason unavailable, and the recruiters were fishing in other ponds, it might be quite possible that wages would be dramatically lower and that the thugs would be much less effective. Of course, another way of saying that there was a radical shift in the supply curve for state violence is to say that there was a revolution.

lawyer

Meanwhile, a scoop from Robert Fisk. This is amusing; if some sort of art-terrorist group had wanted to mock US policy, could they have done better than appointing as special envoy a man whose father was the CIA’s head of planning and who is actually Mubarak’s lawyer? The State Department should probably review personnel policy. It keeps them out of trouble, after all.

Daniel Davies‘s post about arseholes, and more formally about the importance of the reactionary mob as an institution, has been a well deserved hit. Here’s something interesting, though. Fairly serious rumours reckoned that the arseholes were being paid as much as $68 a day. In theory, if an arsehole was on duty 340 days a year, they’d make $23,120 a year (presumably cash in hand, too). Egypt’s per capita GDP for 2010 was $6,200.

To put it another way, when the state needed thugs, it had to pay four times the per capita average income. Of course, it’s possible that these numbers are seriously in error. But the principle isn’t obviously false – mercenaries are usually paid a much higher spread over the typical income of the country where they operate, an implicit recognition of the fact the people want nothing to do with them or those who hire them.

In more advanced markets for thuggery, though, it’s typical to hire someone for a specific act of violence, at rates considerably lower than per capita GDP. What does this tell us?

detail

So what happened in Tunisia? It’s probably worth pointing out that they’ve signed a gaggle of UN human rights conventions, dissolved the old ruling party, and are having a strike wave. Having done the broad strokes of the revolution, they’re now working on the detail.

Is it meaningful to say that the Egyptian revolution is calming down, or petering out? I ask because a common flaw of the reporting on it has been to treat the basic dynamics of mobilisation as if they were signs of huge political shifts behind the curtain. It’s obviously true that both revolutionaries and reactionaries need to sleep and eat. When the revolutionaries want to, they have no great difficulty in putting over a million people on the streets in Cairo and probably a bit more again elsewhere in Egypt. These are peak efforts. Idiot management-speakers like to talk about maintaining peak performance, but they are idiots: the word peak implies a supreme effort that cannot be maintained continuously. People have to eat and sleep, they have families, they have jobs, although many millions of Egyptians have been taking part in the revolution silently by essentially going on strike. Even revolutionaries have to maintain their barricades, update their blogs, and hold meetings to decide what to do next.

The result of this is that there’s been a sort of media cycle – one day the papers are full of pictures from the latest day of rage, the next it’s all about people grandly speculating on what happens next, and the regime’s spokesmen explaining how they intend to preserve the substance of the regime. Perhaps they talk about that on the other days, but nobody is listening. Or perhaps they believe it, when they wake up and hear that there are only tens of thousands of rebels in Tahrir Square rather than hundreds of thousands. Then, the next callout of the demonstrators resets the clock again.

Today, we seem to be in one of the ebb-tide phases. So it’s a good moment for a bit of speculating. What is important, in these terms, is that the government doesn’t seem to be regaining much ground in between waves of protest. Instead, there seems to be a ratchet in operation – each wave extracts a new concession. Mubarak sacked his government. And appointed a vice president. Then he promised not to stand again. Then talks were opened with the opposition. Then the military accepted to talk directly with the opposition, independently. Then the NDP hierarchy was purged. Then Suleiman renounced becoming president himself. And the regime’s own peak effort – Wednesday’s thug raid – was dramatic and violent at the time, but with hindsight was nowhere near enough in terms of numbers to change anything. Arguably, it wrecked the government’s remaining legitimacy and only demonstrated its lack of mass support.

The fear is that this is no ratchet, but a sort of retreat into the Russian hinterland, a trap. On the other hand, it’s a common pattern in the end of dictatorship, a sort of political Cheyne-Stokes breathing. You may think you are saving the structural realities of power and giving away the forms, but how will those realities stand up without the Emergency Law and the special constitutional amendments and the practice of having political prisoners and the ban on opposition parties and the censorship of the press? After all, there must be a reason, rooted in the structural realities of power, why you wanted them in the first place. If owning hotels was enough to sustain a tyranny, there’d be no need for Central Security or private thugs on camels or sententious TV broadcasts or bulk SMS messages with faked originating numbers.

Revolutions come with years, like New Order remixes used to. Prague ’89. Paris ’68. Probably the most relevant ones now are the Polish ones – Solidarity feat. Jaruzelski ’81 and ’89. The first one was a lot like what everyone fears for Egypt and also quite a lot like the official preferences of our governments. There was violence, but not as much as there could have been, and a safe military dictator won. He, in turn, turned to a religious and conservative pseudo-opposition to give his rule some foundation. The second was more optimistic but less spectacular. In 1989, the end of communism in Poland involved far more negotiating than it did street-fighting, and it involved putting up with Jaruzelski sticking around for the rest of his term as a sop to the powers that be, or rather the powers that were.

Egypt is already some way beyond 1981 – there is something like a round table, and the officially designated military strongman is getting very close to the exit, having disclaimed supreme power for himself. Probably the communists of 1989 thought they were cunningly playing for time. Suleiman has a far more ruthless reputation, though; the big issue is whether he can be trusted or better, constrained from trying to either crush the opposition between here and whenever the election date is set or else to start a civil war like the Algerian generals of 1991.

One argument has been that there would be a fake revolution, leaving the security state in charge, as Jamie Kenny put it. I think this is now out of date. Similarly, although they are now talking to the Muslim Brotherhood, I think my own prediction is also out of date. We’re past the point where a few Brothers in the government would convince anyone. In fact, Jamie and I saw our predictions first validated and then rendered irrelevant within a week.

Looking ahead, it’s worth remembering that 1989 took time to deliver. After the original moment of success, there was a long and uncertain haul of getting rid of specific individual bastards, changing laws, moving editors around the State TV and inspectors around the police force. I think we’re now into this phase. Some people seem to agree, from very different points on the spectrum. Changing the union confederation and the university professors’ club is very much to the point, whether you’re thinking 1989 and maintaining enough forward momentum to protect the revolution or 1917 and the second wave.

Take it easy ya Ahmad. Every revolution in history always has this carnival-like side. The insurrection will come later. #Jan25

I think I’d rather have that man on my side.