4 ms·

> The modern Wikipedia hosts 11–12 times as many pages as it did in 2005,but the WMF is spending 33 times as much on hosting This doesn't make any sense to me.

> The modern Wikipedia hosts 11–12 times as many pages as it did in 2005,but the WMF is spending 33 times as much on hosting This doesn't make any sense to me. Why would the number of pages and hosting costs be linearly correlated? I can't think of any other scenario besides Wikipedia where this would be expected to be true.

busymom0 7y ago
Also hasn't the cost of hosting gone down significantly since 2005?
logifail 7y ago
> Why would the number of pages and hosting costs be linearly correlated? How would one expect them to be correlated? As an aside, a friend of mine relaunched an ecommerce site in March. Wordpress + WooCommerce, stripped of fluff. His hosting plan costs him $3/month, including tax. He told me his site took just over $60k in revenue in the first 6 weeks.
- darawk 7y ago
  I assume the operative word for him was 'linearly' not 'correlated'. But cost/pageview is sublinearly correlated, so spending growth should be even slower than a linear projection.
  - logifail 7y ago
    > spending growth should be even slower than a linear projection Indeed. In my experience in several organisations, as soon as there is even an inkling of "spare money" in (part of) an organisation, humans will invariably starting coming up with creative ways to spend it. This may end badly. If you start with the premise: "there is [almost] no budget, what can we do anyway?", then you often get the most creative and useful solutions.
    - Reimersholme 7y ago
      "Parkinson's law is the adage that 'work expands so as to fill the time available for its completion'. It is sometimes applied to the growth of bureaucracy in an organization." Fittingly enough, there's a Wikipedia article on the subject. :) https://en.wikipedia.org/wiki/Parkinson%27s_law https://en.wikipedia.org/wiki/Parkinson%27s_law
  - mastazi 7y ago
    I just wanted to point out that the article mentions the number of pages, not pageviews, which is part of the reason why I find that statement so bizarre.
    - interactivecode 7y ago
      right, because you pay for traffic (or results of traffic) and not just having the pages online.
    - elect_engineer 7y ago
      Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer". See https://news.ycombinator.com/reply?id=21700802 https://news.ycombinator.com/reply?id=21700802 and https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_Cancer https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_C... This is the first time I have seen any comment saying that I should have listed page views instead of pages. I hadn't given it much thought until you brought it up, but I don't see any reason which I shouldn't remove the pages statistic and replace it with page views. I do want to make the essay as accurate as I can, and criticisms/suggestions are a big help in doing that. (you never see the flaws in your own writing). Can anyone think of a downside to me changing it to pageviews?
      - FabHK 7y ago
        > Can anyone think of a downside to me changing it to pageviews? No, it makes sense.
      - Sacho 7y ago
        Sure - I would argue that bandwith costs are only a tiny fraction, and the real cost is in stewarding all the existing content(page numbers) - e.g. more servers = more admins, older content/tools need maintenance, etc. I think what is happening here is that the server costs argument is weak as a first argument. Maybe x33 expenditure is high, but I know WP has grown, and it's hard to quantify how that growth translates into expenditure - it's easy to imagine higher reliability, more accessibility, more need for maintenance of established products and so on as WP matures. If I happen to like WP and am willing to give them benefit of the doubt, why wouldn't I trust that their engineering team knows what they're doing? You haven't established your credibility as an SRE(or enough facts) to really shake that initial goodwill. Once I view your initial argument as shaky, I am naturally more inclined to view the rest of your arguments skeptically(or just outright ignore them). From an outsider POV, your two strong arguments are: - I want to donate to just WP because I don't care about the rest; As the non-WP fraction of expenses grows, there will be a tipping point where the "waste" will erode my willingness to donate - Even though there are growing development costs, WP tools have stagnated, core problems aren't addressed, new features aren't being used(another signal of "waste") I think if you lead your essay with those arguments, the server costs argument would be better received. It's much easier to comprehend "Out of $1 I donate, only $0.50 goes to WP", and as an experienced WP user you speak from a position of authority on the second point. From there, it's easier to convince me that WP is also being wasteful in server costs. In general though, I think you need more evidence put forward to really convince an outsider that WP is wasteful - more comparisons of overall spending vs just-WP spending, more examples of development stalling or going in the wrong direction. The essay may be convincing to an insider who already is aware of the problems, but from an outside perspective it's light on evidence.
sleavey 7y ago
I guess images and videos get larger and more frequent on pages as they are updated.
- smcl 7y ago
  This makes sense - and additionally as higher speed internet becomes more prevalent, it's almost become expected to serve more images and videos and for those to be higher quality. That's a total guess, but in general I'm of the belief that a higher hosting bill does not necessarily imply waste or that something nefarious is afoot
ksec 7y ago
I am wondering, If in 2005 it cost them $5000 to serve 1.4B page views. Could this be done in 2015 with $2500 / $1000 or even $500.

busymom0 7y ago

Also hasn't the cost of hosting gone down significantly since 2005?

logifail 7y ago

> Why would the number of pages and hosting costs be linearly correlated? How would one expect them to be correlated? As an aside, a friend of mine relaunched an ecommerce site in March. Wordpress + WooCommerce, stripped of fluff. His hosting plan costs him $3/month, including tax. He told me his site took just over $60k in revenue in the first 6 weeks.

darawk 7y ago
I assume the operative word for him was 'linearly' not 'correlated'. But cost/pageview is sublinearly correlated, so spending growth should be even slower than a linear projection.
- logifail 7y ago
  > spending growth should be even slower than a linear projection Indeed. In my experience in several organisations, as soon as there is even an inkling of "spare money" in (part of) an organisation, humans will invariably starting coming up with creative ways to spend it. This may end badly. If you start with the premise: "there is [almost] no budget, what can we do anyway?", then you often get the most creative and useful solutions.
  - Reimersholme 7y ago
    "Parkinson's law is the adage that 'work expands so as to fill the time available for its completion'. It is sometimes applied to the growth of bureaucracy in an organization." Fittingly enough, there's a Wikipedia article on the subject. :) https://en.wikipedia.org/wiki/Parkinson%27s_law https://en.wikipedia.org/wiki/Parkinson%27s_law
- mastazi 7y ago
  I just wanted to point out that the article mentions the number of pages, not pageviews, which is part of the reason why I find that statement so bizarre.
  - interactivecode 7y ago
    right, because you pay for traffic (or results of traffic) and not just having the pages online.
  - elect_engineer 7y ago
    Full disclosure: I am the author of the Wikipedia essay "Wikipedia has cancer". See https://news.ycombinator.com/reply?id=21700802 https://news.ycombinator.com/reply?id=21700802 and https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_Cancer https://en.wikipedia.org/wiki/User:Guy_Macon/Wikipedia_has_C... This is the first time I have seen any comment saying that I should have listed page views instead of pages. I hadn't given it much thought until you brought it up, but I don't see any reason which I shouldn't remove the pages statistic and replace it with page views. I do want to make the essay as accurate as I can, and criticisms/suggestions are a big help in doing that. (you never see the flaws in your own writing). Can anyone think of a downside to me changing it to pageviews?
    - FabHK 7y ago
      > Can anyone think of a downside to me changing it to pageviews? No, it makes sense.
    - Sacho 7y ago
      Sure - I would argue that bandwith costs are only a tiny fraction, and the real cost is in stewarding all the existing content(page numbers) - e.g. more servers = more admins, older content/tools need maintenance, etc. I think what is happening here is that the server costs argument is weak as a first argument. Maybe x33 expenditure is high, but I know WP has grown, and it's hard to quantify how that growth translates into expenditure - it's easy to imagine higher reliability, more accessibility, more need for maintenance of established products and so on as WP matures. If I happen to like WP and am willing to give them benefit of the doubt, why wouldn't I trust that their engineering team knows what they're doing? You haven't established your credibility as an SRE(or enough facts) to really shake that initial goodwill. Once I view your initial argument as shaky, I am naturally more inclined to view the rest of your arguments skeptically(or just outright ignore them). From an outsider POV, your two strong arguments are: - I want to donate to just WP because I don't care about the rest; As the non-WP fraction of expenses grows, there will be a tipping point where the "waste" will erode my willingness to donate - Even though there are growing development costs, WP tools have stagnated, core problems aren't addressed, new features aren't being used(another signal of "waste") I think if you lead your essay with those arguments, the server costs argument would be better received. It's much easier to comprehend "Out of $1 I donate, only $0.50 goes to WP", and as an experienced WP user you speak from a position of authority on the second point. From there, it's easier to convince me that WP is also being wasteful in server costs. In general though, I think you need more evidence put forward to really convince an outsider that WP is wasteful - more comparisons of overall spending vs just-WP spending, more examples of development stalling or going in the wrong direction. The essay may be convincing to an insider who already is aware of the problems, but from an outside perspective it's light on evidence.

sleavey 7y ago

I guess images and videos get larger and more frequent on pages as they are updated.

smcl 7y ago
This makes sense - and additionally as higher speed internet becomes more prevalent, it's almost become expected to serve more images and videos and for those to be higher quality. That's a total guess, but in general I'm of the belief that a higher hosting bill does not necessarily imply waste or that something nefarious is afoot

ksec 7y ago

I am wondering, If in 2005 it cost them $5000 to serve 1.4B page views. Could this be done in 2015 with $2500 / $1000 or even $500.