LJ to restructure backend, redesign interface, eliminate advertising?

[charmian]

http://translate.googleusercontent.com/translate_c?hl=en&rurl=translate.google.com&sl=ru&tl=en&twu=1&u=http://lenta.ru/news/2011/11/19/zhzh/&usg=ALkJrhhxg9B2Rg2k0WsNodEgzwrk3iiL9A <--Translation of a news article claiming LJ will not be using ads (however, users will be allowed to put their own ads on their journals).

http://translate.google.com/translate?sl=ru&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Flenta.ru%2Fnews%2F2011%2F11%2F19%2Fzhzh1%2F <--LJ will create a revamped system for ranking bloggers, update the site design (wow), described as being unchanged from 1999, and the backend will be rewritten, allowing uses to have more friends. Now, since this is all via google translate, I'm not sure how accurate this is (anyone who reads Russian want to confirm this?)

Servers may also be moved from the US to Europe (I've seen other sources saying some, but not all will be moved).

http://translate.google.com/translate?hl=en&sl=ru&tl=en&u=http%3A%2F%2Faspix.livejournal.com%2F133821.html <--Better picture of new interface, and mention that there will be more extensive use of Ajax.


Huh, if the servers are moved to Europe, what consequences will this have for US users?

Where are they getting all the money to pay for this? (Reduce/eliminate ads, new design, redoing backend)

Is there going to be much disruption to service if they're going to redo the backend?

What consequences will more Ajax use have?

I wonder what people will think of the redesign of the site pages.

Comments

[acari]

Do you have any idea where the servers will move, which country or even EU or non-EU?

[charmian]

I don't think they specifically said.

[synecdochic]

Huh. This reminds me a lot of the plan that has been floated on and off since I was still working for them: rewrite the engine in order to close the source. (It could be done: if it's rewritten all at once, instead of piecemeal, with all the work being done by SUP employees, they would not have to keep the GPL.)

I'm curious to see how the 'karma' works, especially since Brad tried to do that a while back (when invite codes went away) and got, well, let's just say "some pushback".

I'm also curious to see how they intend to change the backend friends code...

[charmian]

By "in order to close the source" you mean for the purpose of closing the source? Is that the motivation here? What benefit do they get from closing the source?

Oh, they already have a system like that in place, I think. They are talking about a revision to make it more based on social, which is why they were saying the ratings were going to change, but not that much?

Yeah, it sounds like quite a change in the friends stuff. I wonder how they are going to let you have 10K people as friends (who can read that many journals?)

[fiddlingfrog.livejournal.com]

What consequences will more Ajax use have?
It'll further lock out people who refuse to or have difficulty using JavaScript. You can already see some AJAX in use in the ScrapBook II beta test The post an entry page was recently redesigned to use JavaScript in the date widget, in anticipation of some upcoming features, and there have already been some user complaints.

Huh, if the servers are moved to Europe, what consequences will this have for US users?
My bet is that they're somehow going to move Cyrillic-service users to a European server, for faster access and load times, and possibly leave other users on US-based servers.

As for page design, I'd expect to see a change to http://www.livejournal.ru first before they get around to http://www.livejournal.com. If you look through the changelog you can also see that there's going to be more changes to http://www.livejournal.sg in the near future.

[charmian]

Huh, I find LJ already pretty darn unusable without Javascript. What features are they preparing that will use the date widget?

Huh, are they already separated like that?

That will be pretty interesting. I guess I'll keep an eye on that. It seems the entry pages will be further redesigned, and perhaps the commenting pages, although what I am most curious about is what exactly they're going to do with the backend.

[fiddlingfrog.livejournal.com]

Nothing official but looking through the changelog it looks like draft entries (write now, have it post automatically later) and permanent sticky entries without the "date out of order" trick.

I don't think so. I don't even know if it's possible, but moving Russian users to a server that's closer to them geographically is the best way to get the "faster service for Russian users" that I've seen igrick talk about.

I think I've seen something about commenting changes in the changelog, but it's harder to get a read on what's being changed without a deeper understanding of the code than I have.

[charmian]

Ah, I see.

Yeah, I have no idea how it's done. However someone told me that moving it closer to Russia wouldn't necessarily give that many speed improvements?

I've also seen commenting changes, but don't know anything about coding, so I have no idea either.

[sophie]

"My bet is that they're somehow going to move Cyrillic-service users to a European server, for faster access and load times, and possibly leave other users on US-based servers."

That's very interesting - presumably it means that the Russian government won't have control of the servers. If I remember rightly, there was some uproar about that before.

[elena]

Huh, if the servers are moved to Europe, what consequences will this have for US users?

We have better data protection/privacy laws, to start with \(º3º)/

[charmian]

That sounds nice!

However, I have never really understood why it matters so much where the servers are. If you can't trust the owners of the site, it doesn't really matter where the servers are, I think.

[azurelunatic]

The servers are less vulnerable to government agents knocking politely or not-so-politely at the data center when the data center is not in the same country as the agents who would like to inspect the servers.

[charmian]

Indeed, but isn't the data center already controlled by the headquarters of the company? In other words, so if the HQ is already in the country whose government you're worried about, isn't it sorta game over already?

[jd]

Well, yes and no. I mean, if they're physically outside the country, the government can't just go to that country and seize them. It's at least another layer of difficulty, is how I see it.

[azurelunatic]

There were protests from LJ users in Russia when SUP started handling stuff at all, and again when SUP took over. So I think that's a significant point of vulnerability.

A hypothetical hostile agent could demand that the company surrender records, or take possession of employees' computers, and demand access. Certainly. And with digital data a copy is usually the equivalent to the original.

Remote access has two ends; I don't know what would happen if someone took an LJ employee's computer, but if I were a paranoid boss I might suspend an employee's accesses if they went missing, particularly if the things they could access were particularly sensitive. I might also require possession of a physical token and regular entry of a password known only to the person who ought to have that token, and automatic suspension of access following failed authentication. Basically there are steps that people can take to secure data from unauthorized access.

It is more difficult to defend your data when the device on which it is stored is in the hands of hostile agents. If you've deleted something by erasing the pointer to the place on the disk where it was stored, physical possession of the disk can lead to restoration of that data. If something's been overwritten with zeroes, sufficiently sensitive tools can pick up magnetic shadows on magnetic media. This is why modern data destruction involves physically destroying the drive.

Defense of data isn't the only thing; persistence of service is another vital factor. Pinboard went actually down at one point because the FBI took the server. Pinboard was on the same physical machine as the virtual machine that some entirely un-fun things were located.

When the goal is not to find out information, but to make information inaccessible or excessively inconvenient to retrieve, physical access makes this much easier. DDoS only makes it inconvenient for a while, and does damage to business. There's only so much that a person could erase remotely before getting caught and cut off. If you really want to destroy others' access to data, you get hold of the main server and as many backups as you can, and you screw with that.

I do not imagine that that would be a subtle process.