Dave

Just to update everyone on this, the new server is now in place and operational. We've already transferred the majority of services over to the new server. Our plan is to transfer SFN itself on Saturday 15th February, at around 4pm GMT. In all likelihood the downtime should be relatively minimal, hopefully no more than 1 hour.

Many apologies to our European visitors for the extended downtime this morning. Unfortunately, our database backup scripts caused us to run out of memory on our machine this morning, making it unresponsive and requiring a power cycle. The subsequent loss of power caused a further delay as the RAID array required resynchronisation. However we are now back online!

Clearly, server issues have become quite the problem over the last few months, which is almost entirely due to increasing memory requirements as our forum grows and the software driving it becomes more complex. For various reasons of both practicality and finance, the easiest way to alleviate this problem is to replace our 7-year-old server with something which is a little more up-to-date and will provide further room for the growth of SFN.

I'm therefore pleased to say that this process has begun, and over the next week I should be able to update you on when the new server will come online.

In the meantime, please bear with us and look forward to a snappier SFN!

Okay, a brief update from me on this.

We've identified a couple of potential areas of concern, mostly to do with the database backend which seemed to be under quite heavy strain. We use the SFN server for a few other websites, and one of these seems to have a table that was excessively large. This has been truncated now, so we'll see if the database server has some relief now! There has also been an update applied to PHP (which drives IPB, our forum software) which has also been applied. Plus we are tweaking quite a few parameters to optimise memory usage generally on the server.

The state of the forum is very important to me - after all I've been knocking around these parts going on 10 years now. Ultimately if the forum continues to be slow over the next 1-2 weeks, the problem can be rectified through some hardware upgrades to the server. If we need to do this then we will!

Apologies, guys. I've also noted the server is extremely slow as of late, hence my initial modifications to try to tune things which clearly failed! I'll try to convene with Capn' and sort this all out over the next few days.

Are these updates going to help in with the bots in any way, shape or form?

I missed this one, sorry. Hopefully, yes. But can't promise anything! I'll be looking into other options as well, obviously this is a big priority for us at the moment.

I've refreshed several times, but don't see the reputation buttons.

 

Ah, now I see them.

 

And now I don't.

You might try clearing the browser cache?

so far everything went back to normal on my end.

Great!

Good work dave. I'm sure most of us, maybe all, have been on the receiving end of Murphy's wrath.

 

I'm pretty sure no one had anything worse happen to them than I, which was severe withdrawal and slipping into a coma (err falling asleep) last night during the worst of it.

 

Whatever you live through makes you stronger.

Thanks Ed

Hi guys, my apologies for the issues. Unfortunately there were some critical security updates which needed to be applied, otherwise I would have notified in advance of the updates. Combined with a few family issues over the last 24 hours this hasn't been as smooth as I would have liked.

Hopefully the offline message is now sorted (a bit of a web server issue). I'm working on correcting the rest of the issues and will have an update for you later on this evening.

Update: Okay, I've pushed a few buttons and the style issues seem to be resolved. But if anyone still has problems, let us know - just be sure to do a refresh or two first.

Hopefully this issue should now be resolved (or at least very much improved), but please make sure you report any suspected spam!

You will probably have noticed that following a couple of hours upgrading, we have successfully upgraded to the latest version of our forum software. The reasons for upgrading are pretty numerous and there are some nice new features, including changes to make the editor easier to use, improvements to the search engine and some performance enhancements. However, we hope that by far the most marked impact will be a reduction in spam messages throughout the forum.

One thing that you will almost certainly have noticed is that the site is looking quite a bit different! Unfortunately the changes to themes have meant that we have had to remove quite a lot of our customisations. We hope that, over the course of the coming days, we can restore some of our old looks!

A disclaimer that I haven't done topology in a good few years now, but whilst it's true that properties which are preserved in a topological sense are important and useful, techniques involving gluing or cutting can be used just as well too. Topological constants aren't everything

Rainbow tables are useful for reversing hashes, but they're rendered fairly inert by use of a salt which is stored with the hash. Wikipedia has a pretty good entry on them.

I think that they are making the observation that r is very large (and much bigger than [math]R^2[/math]). Therefore the first time is very close to zero and thus ignored.

Honestly, if I were you, I would look into taking a basic course. Looking back on when I taught myself some more advanced topics throughout school, I found that I wouldn't spend enough time truly understanding certain aspects and had a tendency to move too quickly through a syllabus. Having a course gives you a set structure to work with, and allows you enough time to absorb and understand the material.

Anyway, good luck in any case!

Hah, fair enough - it's not often you get a lawyer posting to these kinds of things so I will bow to your superior knowledge of this

Victor, at this point I think we can agree this isn't really mathematics and has been dragged on for a long time now. Therefore, I'm moving this thread to speculations. Please do not attempt to start any new threads on this topic in the mathematics forum; they will either be locked or deleted on sight.

My interests are on completely the other end of the scale to you guys. For my PhD I've been focused on studying fluid dynamics (particularly flow through a pipe), and really the mathematics comes into play in terms of numerical analysis and getting decent schemes to solve Navier-Stokes. So at this point I guess I'm somewhere between a physicist, engineer and mathematician.

The thing is that I quite like that. The idea that I could focus purely on, say, analysis or topology and not be able to apply that to something which is a bit more physical isn't the kind of thing that I can easily get into, and is certainly not anything that I would enjoy.

So I have to take umbrage with your last comment DrRocket. In this (mathematics) department there is no shortage of 'applied mathematicians' who are in themselves excellent at mathematics and know the applications of their specific subfield inside out. It's worth noting that I don't consider myself to be one of them and I've always been more on the physics side of things, but the point still stands regardless.

I don't think this is where you were going with this, but to air a particular grievance of mine anyway: I really detest the snobbery that both pure and applied mathematicians seem to have for one another. Personally, I have the utmost respect for anyone that chooses to study mathematics (and the related subjects) at a higher level. I'm very grateful to work in a department which actively encourages collaboration, and this shows in the quality of the research that gets put out. At the end of the day, pure mathematics is pretty much useless to the wider population without applications. Likewise, applied mathematics relies on theoretical results in order to progress. Since it's pretty much a fact that the level of funding is largely reliant on the usefulness of the applications that come out of the work, in my mind at least there is nothing to lose by collaborating and an awful lot to gain.

These codes are not trivial and require a lot of mathematical analysis. A starting point may be to consider the papers that have been published in and around the area. However, it is incredibly unlikely that anyone here (including myself) will be able to assist; even if they could, I have no doubt that they would question your motives for doing this.

I disagree with your post slightly imatfaal, in that even though IANAL, I would believe that it is almost certainly not illegal to find a collision for your checksum (at least in the UK). Obviously, if you then use this for any sinister purposes you will potentially find yourself in a lot of trouble.

I tend to agree with Pangloss on the poster, since it's not clear at the moment that it motivated the crime at all. That being said, if it turns out that it is this guy's desktop background and it's printed all over his apartment, then clearly it will be an issue.

I think the point that most people are trying to make is that the poster, at least for the moment, is symbolic of the mud-slinging and hateful comments that are thrown about by representatives at both ends of the spectrum. That being said at least, the right-wingers in the US are, from what I can see, far more prolific and hateful than the left-wingers. Case in point:

http://obamalondon.blogspot.com/2011/01/inexplicable-edits-on-sarah-palins.html

This is exactly the kind of thing that I hate, and I find it incredibly disrespectful. Clearly, Sarah Palin's team hasn't changed their attitude much towards this stuff yet.

Statistics is a subset of mathematics which is devoted to understanding the notion of chance.

As far as I know there's no particular "advantage". I guess if you had proven some result that used this scaling then it could be handy. Why do you ask?

I'm closing this thread - quite frankly the posts make no sense.

Looks fine, as long as you mention it's all on average

Hi zheng,

Thanks for taking the time to post, but if you just want to say thanks for posting the question it's customary to give the poster some reputation (or karma). You can do this with the little plus button at the bottom right corner of their post.

Thanks,

Dave

Dear all,

I have recently updated the forum permissions to require members to have 30 posts or more before they are able to reply to threads or create new threads in this forum. Since SFN is a primarily science forum, this is a requirement we have on a variety of the off-topic fora.

Thanks,

Dave

You did not answer my question. Scroll back up to #33, stop dodging the question, and answer. Or your lack of effort will again out you for coming here to compete with people, instead of arguments.

 

Marat is very easy to follow. Just read. You appear to be calibrated to analyze a forum member, and their belief system, but public discussion boards are not about analyzing posters, but rather their arguments. One's ego often is responsible for such things.

 

When you remove the 'person' from the 'argument', you are left with multiple arguments. Marat's posts are essentially several different arguments - all are up for discussion and debate. Pick one and argue it. Or pick several. Hey, if it helps, make up little names for each argument and pretend a non-existent member posted them.

 

There is no functional problem with arguing multiple angles of a subject - except, if you're aim is to compete with the person making the argument.

ParanoiA, if you have a point to make then do so in a civil fashion. Insulting someone by inferring they have a big ego is not what is generally regarded as reasoned debate.