Ask MetaFilter questions tagged with optimisation

How can I make my rented flat warmer? On the cheap, if possible?

handee — Wed, 14 Oct 2009 15:58:31 -0800

I'm renting a flat and it's cold, and it's going to get colder (I'm on a 1 year contract in the alps). Is there anything noninvasive (preferably cheap, but I'm ok to spend if it's really going to work) that I can do to warm the place up? What sort of portable space heaters give the best performance for electricity/gas? Are there any decent renter-based guides to improving insulation? There's one mains gas heater in the middle of the flat that's supposed to cover the whole thing. I have shutters and decent single glazing. Local people who've visited have looked at the heater and laughed.

How to maximise a matrix subject to constraints?

muhonnin — Sun, 14 Jun 2009 05:16:45 -0800

Optimisation problem: trying to find the subset of my data that meets a couple of constraints. I have a data set with 200 columns and 20 rows of numbers. I am trying to find the 20 columns that make the mean of all values in the resulting 20 x 20 data set as large as possible--subject to the constraint that no row in the 20 x 20 set can have a mean value below a particular figure.

I have access to a range of software tools (Excel, Mathematica, SPSS). I know how to specify the constraint if I were going for a brute force solution, but obviously that's not going to be feasible here. All suggestions greatly appreciated. . . . .

What to do with spare download bandwidth?

kureshii — Sun, 01 Jun 2008 22:57:09 -0800

Is there anything meaningful I can do with 10Mbps worth of spare download bandwidth? My ISP recently has an offer on a 10Mbps connection, under a 2-year contract. I've been thinking about getting a separate connection, mainly because the price is good and because I want a good upload bandwidth for torrent-seeding (don't ask).

That plan comes with 1Mbps upload bandwidth which, as mentioned, will be largely used for torrent-seeding. That leaves the 10Mbps download bandwidth largely unemployed. I will still be downloading every so often, but hardly enough to put more than the occasional dent in it.

This is where I need to pick your brains. Are there any meaningful projects out there requiring spare bandwidth, and if so, how can I contribute? Is there are other way I can put it to good use?

I've also been thinking about social network-related experiments and data-logging webcrawlers that would gather historical, social and other data (of course, ideally without maiming other servers), but have yet to think of a meaningful way to turn that into useful insights or conclusions.

If anyone has ideas on what to do with that download bandwidth (that would preferably leave my upload bandwidth alone), please hit me with them =)

How is Facebook doing its queries?

bonaldi — Mon, 04 Feb 2008 09:49:41 -0800

How does Facebook handle or simplify the presumably complicated DB queries involved so that me loading my page doesn't bring it to its knees? When I pull up my homepage on Facebook, it pulls status updates for a subset of my friends, and recent bits of their activity. It filters out the ones that are related to people who have blocked me (hopefully none!) and then shows it all to me.

This information is all very specific to me, so presumably caching won't help much with it. But it seems like a hell of a lot of DB queries. If that's so -- it may merely be my "SELECT *"-level understanding of SQL making it seem that way -- how do they do it efficiently?

(I'm particularly interested in how the 'blocked user' can work: would it really pull a status update like "Matt Haughey and Jessamyn West are now friends", then pull Jess's block list to see if I'm on it? That's two hits for a one-line entry!)

Sock quality modelling and optimisation?

TheNewWazoo — Thu, 14 Jun 2007 07:10:01 -0800

Let's say I buy s socks. Now, these socks feel the very best (have a quality factor of 100) after f wearings, and must be discarded (have a quality factor of zero) after f × 100 wearings. I want to add these socks to a pool from which I draw two daily, and want to maximise the probability of selecting socks with the highest quality over the life of the pool... So, presuming I want a function which describes the quality of my socks... how do I model and optimize it? I've taken a few courses in basic calculus and statistics, but have no idea how to model this. What branch of mathematics deals with these sorts of problems, and what is a good (better?) example of an attendant/canonical problem for that branch?

This is not homework!

Do they shape up, or do I ship out?

lemonfridge — Wed, 21 Sep 2005 03:57:12 -0800

I just started working for an IT company that has been dipping it's toes in the world of web design & development for a while, but not been doing things right. How do I get them to shape up? Or do I ship out? Sites by the other programmer are mostly IE only, and used ASP and Access Databases. Which says it all. I'm trying to move things into XHTML and CSS, and using PHP + MySql, but I'm getting a lot of resistence. I know it's a better way of working, but I have trouble explaing WHY. I respect that the other programmer may have to learn new things, but I'm also being expected to learn ASP, which I am.

I'm also being asked to provide search engine optimisation, which I've always considered selling snake oil. I know there are some things you can do, but I don't expect them to make much of a difference.

Should I try to change them and stand up for what I belive is right? or just buckle and do as I'm told (and be unhappy)?

I've only be here 3 months, but I feel the urge to leave already. It has a lot of promise (my own office and working from home), but I'm being forced to bodge jobs that I would never do elsewhere.