This is a clear violation of net neutrality principles, but it's also clearly a good thing for the Iraqi people. How do those in the hard-core NN tent square that circle?
To sidestep any straw-man arguments about war zones, different country, etc; What if this were implemented in Appalachia to give access to wikipedia to the poor an uneducated there?
The hardcore NN tent, at least on HN, isn't particularly hardcore, and especially not about NN. They are more hardcore (and, for that matter, more reasonable) about wanting to cripple (or in some cases "regulate") big communications providers who use their monopolies in the long-distance bus topology space to push their own media agenda. And NN happens to be a politically viable tool for that.
But if you put it to a vote, a more anarchic internet (characterized by, inter alia, 1. wholly distributed naming and cert mechanisms, 2. final-mile mesh networking, 3. few or no blanket bans on low-power radio transmission, 4. the removal of politically-enshrined content bias from the application layer (ie, copyright law and software patents)) will, I think, win such a poll. The overwhelming majority here prefer that vision as a long-term plan, compared to barely-leashed telecoms restrained by a wink-wink-nudge-nudge "net neutrality."
And so, the solution referred here, since it is pragmatic and reasonable, meets a sort of "for now" test of decency, whilst not being inconsistent with a soft, "best we can do" system of net neutrality for Comcast and Time Warner.
> but it's also clearly a good thing for the Iraqi people
Wouldn't it be even better for the Iraqi people if they could also have affordable Internet access for the purposes of communicating and conducting commerce? Unless Wikimedia is footing the bill for the data usage, this is just a case of the telcos restructuring their pricing so that the lost profits of Wikimedia access are subsidized by the rest of the data usage. At least some of that other usage is probably more valuable to the Iraqi economy and general welfare (especially in the near term) than reading Wikipedia, even if it is less idealistic sounding.
The two are not mutually exclusive. If the marketplace is competitive, then ISPs are incentivised to both offer freebies and to make their services affordable.
The problems with the United States are the large area to cover, combined with a relatively uncompetitive market, and the predominance of post-paid contracts. But you can't generalise this to the rest of the world.
> The problems with the United States are the large area to cover,
That argument is never valid. Parts of the US may present a challenge due to low density, but the overall geographical size is not a determining factor for anything relating to internet infrastructure.
NN is about preventing one competitor from getting an unfair advantage. So unless you view Wikimedia as having a competitor, or some future potential competitor, this doesn't really violate any NN principles.
In essence, if we're all OK with Wikimedia having a monopoly on what it is that Wikimedia does then there is no problem.
Going back to the roots of the ICC(predecessor to the FCC) trains could carry USPS mail at different rates than other cargo. The USPS had a government sponsored monopoly on mail delivery. However, rate discrimination between similar cargo from different parties was prohibited. This is the basis for our concept of common carriers and NN.
So your take is that NN regulation should include a test for existing competitors, potential future competitors, and the general agreeability for an entity to become a monopoly?
NN is a pretty loosey-goosey principle that mostly just means "I don't like Comcast". There's also some partisan political football mixed in, and in the EFF's case, sizable donations from content providers like Google who want to preserve a strong bargaining position when buying bandwidth.
Not merely "I don't like Comcast", though I (and a huge portion of their customers) really don't like Comcast.
It is the very high potential, one might even say certainty, that if given the opportunity to hobble their competitors in the entertainment space and to give themselves a massive advantage, they will do so.
The telcos and cable operators in the US have never been anything but monopoly or duopoly players. They don't know how to compete on any other field. If they are given the gun with which to shoot Netflix (for example) in the head (or demand ransoms that only major players can pay), they will likely pull the trigger the same day.
So, no. It's not about disliking Comcast, it is about knowing with confidence that the internet will be harmed, maybe drastically, by giving Comcast (and a few others who play according the same rules) the ability to do so.
Isn't net neutrality bigger issue is things like Netflix paying Comcast to be in their unmetered section - with the very real fear that small websites will get ignored?
I don't put this on the same scale. I'd seriously not mind actual educational websites to be exempt from data charges - and this would include (but not be limited to) Wikipedia. I understand this would have a few flaws and arguements, but it is a start. I'd happily add in direct mostly text-based communication as well (Messenger for facebook but not the pages, hangouts, and the like).
If Netflix just pays for bandwidth, that's net neutrality. If Comcast discriminates pricing based on anything besides byte delivery cost, that's not neutrality.
This is full of holes, but I'll take a stab at this answer.
First, the obvious straightforward sort of information. Wikipedia, for example. Medical advice and self-care, including text-heavy information on cooking, car and machine repair, and so on. Math, science, programming, parenting, history. Primary or secondary language learning and improvment. Anything that actually teaches you a skill (yes, this will be abused to a point, but I'm willing to allow folks to learn to juggle if others can learn to practical skills).
Things like religious texts becomes a tricky area. I think it is good for society when folks learn about different philosophies and religions. Yet these would paid, as well as a lot of the grey area stuff because you gotta draw the line somewhere. Many books fall into this category, even old classics, because most books are intended for entertainment. Some things would depend on context. For example, meditation. In a religious context, no. In a psychology, science, or medical context? stick it in there. Social media is a paid product. Sometimes the cut-off won't make sense: Learning musical instruments would be free, but listening to them paid.
In addition, the site itself would have to take action to be included. If they don't apply to be exempt or to renew exemption, folks must pay for the content.
Net neutrality was a bogus hocus-pocus in my opinion to begin with. Good for Iraq but bad for us ? How and why ? The underlying principles seems completely arbitrary. People should be able to get more access for lesser price and that is always a desirable thing.
You kid, but as much as I like Wikimedia (and as much as I would like to access Wikipedia free of mobile data charges), I think this objectively makes it harder for projects in the same field to compete, even if just a tiny bit.
I like to think I'm a net neutrality supporter, but when my mobile plan included unlimited data for the carrier's own music streaming service, I started using it and the unlimited data came in really handy. I used other streaming services much less because of that; Spotify, for example, didn't stand a chance of seeing my money.
I guess you can't have your cake and eat it too...
As much as I like competition, I think Wikipedia (in terms of Wikis) is as close to Google (in terms of search) as it can get to having a monopoly. And I am more OK with Wikipedia having a monopoly than Google since it is a non-profit. The closest true competitors to Wikipedia are Everipedia, Infogalactic, and Wikia (sort of..kind of) and they all have a fraction of a fraction of Wikipedia's total traffic combined. I am fairly active in Wikipedia's editor community and although there are some issues in the community and real toxicity sometimes in there between different editor factions and deletionists etc, I have met very few people who think Wikipedia has a net negative impact on humanity. It is truly a force for good, so increasing their impact on impoverished areas of the world can only be a good thing.
Sure, but the barrier for the ISP to allowing the same thing for a proprietary music streaming service has been lowered.
This is, apart from the utility that it brings, advertisement for the operator:
> Wikimedia recognizes the user is on that operator's network and serves a banner on the top of the page indicating free data courtesy of their mobile operator, which reinforces a positive brand experience for the operator. [0]
It's not inconceivable that the ISP would do similar deals with $music_streaming_service. Maybe for some payment.
> think this objectively makes it harder for projects in the same field to compete, even if just a tiny bit.
I think this is true in theoretical sense, but not in any practical sense. Just as it's true that you breathing reduces amount of breathable oxygen around, but that's not a ground for me to sue you for air pollution and depriving me of vital oxygen ;) There are things so tiny as to consider them as if they weren't is to seriously distort the matter, and I think it's one of such things.
To make it substantial, one needs to point out that a) there are viable competitors b) they are influenced by things like bandwidth costs and c) this difference is causing real and noticeable preference shift.
I can believe it's true for mobile music streaming services, but I don't see it being true for Wikipedia.
There's one more point - the argument against NN goes "it'd kill Wikipedia Zero". I don't think saying "well, so what, but it'd help Spotify" really a strong counter to it - if I had to choose between living without Wikipedia and living without Spotify, I'd choose Wikipedia without even pausing to think for a second. So NN proponents have to offer better answer than that, if they mean to go the utilitarian road.
> I think this is true in theoretical sense, but not in any practical sense.
I've no idea what current state is, but at one point the Wikimedia Foundation was seriously concerned about potential Wikipedia forks, such as by Facebook. And was working to make them more difficult.
Given that WP seems to have plateaued in a less-than-ideal place, with for instance, a lot of research programming languages considered non-notable, I had mixed feelings about that. A competitor with a different culture could be a good thing.
NN, as I understand it, is not so much about the current state of things, but about making sure that future "wikipedias" and "facebooks" get the same chances as the current ones. Right now there are no viable alternatives (that you and I know of, at least) - but what about this time next year?
You can't change what NN means on a whim, to account for whether each specific type of internet service has "viable alternatives" or not. And how do you even define what are viable alternatives? What's viable for me, might not be viable for others.
In the list of the hurdles you need to overcome to get viable competitor to Wikipedia, zero-rating in Iraq wouldn't be even in the top 20, I imagine.
> You can't change what NN means on a whim
I don't want to change what NN means. I want to emphasize that NN means that people in Iraq who now can have access to a snapshot of the world's knowledge absolutely free wouldn't have that option anymore. In service of some abstract idea of fairness that doesn't even apply to anything specific. I don't think it is a good idea.
> I want to emphasize that NN means that people in Iraq who now can have access to a snapshot of the world's knowledge absolutely free wouldn't have that option anymore.
Not free. The costs are being covered by increased pricing on the rest of their phone usage.
Is there any evidence to support that? Did the phone usage prices in Iraq really surge after introducing zero-rating for Wikipedia? How much did they surge?
My fuzzy recollection is Facebook was serving mirror of Wikipedia, with extensions. Perhaps community pages? https://creativecommons.org/2010/04/21/wikipedia-on-new-face... (2010) The Cooking example there no longer has a copy of the WP page. Perhaps the mirror no longer exists?
In my opinion zero rating is one of those things that is good until it is abused, and it would be much less of a problem if there was actual competition between ISPs. In my opinion net neutrality is largely to keep monopolies (either a web company or an ISP) from gaining even more power over the market. I would prefer to have no zero ratings at all because having any opens the door to abuse but practically it isn't actually bad in this situation otherwise.
Other sources of information can't compete, and it's fairly easy for one source to be controlled by some person or organization. Additionally, when this has been done in the past people turn Wikipedia into the normal internet, hiding media on various pages and using it as a social media stand in.
The money would be better spent lowering the normal cost, but then Wikipedia doesn't get the huge supply of editors.
It's easy to tell other people where their money are best spent, but the situation as it is is that in these countries, the prices of data vs. the salaries are such that 1G may cost you a month's salary. In that situation, the choice is not "listening streaming music on that or this service", it's "zero-rated sites or no internet at all". I'd rather these people had at least Wikipedia access. Even if it may put some Silicon Valley startup in a tougher spot.
Or the money spent providing Wikipedia data could be used to provide a limited amount of free access to the entire internet. Or on better infrastructure so service doesn't cost a months salary.
No, the best solution is free or affordable access to the entire internet. When the majority of users can't read the sources or access new ones to change things, Wikipedia can become a dangerous tool. The English version has problems with political groups controlling what goes on various Wikipedia pages, this system ensures it will happen.
You can't possibly mad for people gaining access to knowledge. For me personally, it's the single most important thing Internet gives.
But as for net neutrality? Of course I'm sad. How are they going to check the sources of the articles? Wikipedia is still pretty limited in some areas of knowledge.
Nothing is perfect. I'm very happy they get to browse wikipedia. It is magnitude of order better than not having access to it.
I know, it's a travesty that we allow Iraqis to access Wikipedia for free. We are literally ending wiki competition. Simply awful. Iraqis must demand net neutrality and stop accessing Wikipedia for free!
Laws and regulations won't be made per site. Net neutrality regulation would be effective only if you don't leave in backdoors via exceptions.
Also, Wikipedia, being text-heavy and without obtrusive ads, is not something for which zero-rating provides much relief.
That sounds dangerously like "we have this shiny new law that sounds good in theory so we're going to ignore that it hurts actual people in practice, because the theory sounds nice". I think if the law doesn't work "per site", it should be rethought until it does.
What has, and where? All it takes is a quick trip to T-Mobile's website to see the list of video services they've zero-rated without any regulatory issues at all.
non-profit just means there is no corporate owner. A non-profit can still be run in a way that enriches management and employees, and a for-profit can have a strong social contribution.
Although zero-rating Wikipedia in itself is rather benign (it has already outcompeted other encyclopedias anyway), it does run the risk of creating more acceptance for zero-rating of other, possibly commercial services.
I am not enraged but this is not good. We don't want ISPs to use their monopolies to pick and choose the web's winners and losers. Just because Wikipedia is a great nonprofit resource doesn't mean that competitors should be disadvantaged.
ISPs don't zero-rate because of the goodness of their non-existant hearts. They do this to get people hooked on using more data to get them to buy more data. Either that or they can use it to start competing services or as blackmail to get web services to pay them.
> Either that or they can use it to start competing services or as blackmail to get web services to pay them.
Who is being blackmailed here? Can you point out somebody who is really hurt by this arrangement, not in theoretical conspiracy scenario but in real world?
In Wikipedia history there was a famous fork of es.wiki called Enciclopedia Libre Universal en Español, which split in some sense over policy disagreements. It remained larger than es.wiki for over two years following the fork.
Right now there are collaboratively-edited encyclopedias in some languages (at least Chinese) that are larger than Wikipedia, like Baidu Baike, which is much larger than zh.wiki.
There were at least two famous attempts to compete with English Wikipedia over policy disputes (Citizendium, which was concerned about verifiability and anonymity, and Conservapedia, which was concerned about the neutral point of view policy). These competitors haven't been very successful, but I can easily imagine people wanting to give each of them another go.
Right now there are about 100 online encyclopedias, each of which potentially competes in some degree with Wikipedia in some sense.
An example that's not particularly a result of an editorial policy dispute is Eric Weisstein's World of Mathematics, a super-great online math encyclopedia (that competes with Wikipedia for attention in its coverage of math). Another is the Stanford Encyclopedia of Philosophy, a super-great online philosophy encyclopedia. Both of these encyclopedias are stronger in at least some parts of their respective areas than Wikipedia is, and perhaps better even on Wikipedia's own terms (e.g., I bet the Stanford Encyclopedia of Philosophy manages to have a more "neutral point of view" on several philosophy controversies than Wikipedia currently does). But zero-rating probably means that people interested in these topics on some mobile carriers are more likely to look only at Wikipedia's articles, even if they're not the best online encyclopedia articles in those areas.
Edit: Eric Weisstein's World of Mathematics has been renamed to "Wolfram MathWorld". Although Wikipedia's coverage of most math topics is really excellent, I still think MathWorld has areas where it's likely to be even better.
Edit 2: Suppose you're interested in a particular religion's official take on some religious issue. Then an encyclopedia edited by adherents of that religious tradition (like one of https://en.wikipedia.org/wiki/List_of_online_encyclopedias#R...) may be more helpful to you than Wikipedia, for example because it may go into more theological or ritual detail, as well as more detail about polemics or apologetics from a particular religious view. That might actually be what you were looking for, but currently none of those religious encyclopedias is zero-rated by a mobile carrier.
Great point, but Baidu is super-rich, so it's unlikely zero-rating Wikipedia would make any effect on them. Not to mention zero-rating in Iraq probably not exactly relevant for zh.wiki :)
> Right now there are about 100 online encyclopedias, each of which potentially competes in some degree with Wikipedia in some sense.
In "some" sense, surely. To be really on the level comparable to Wikipedia by breadth of coverage and participation? Unlikely. None of them would be noticeably impeded by zero-rating Wikipedia in Iraq - most of them probably get next to zero participating from Iraq anyway. Same probably applies to any other place where Wikipedia Zero works.
> but currently none of those religious encyclopedias is zero-rated by a mobile carrier.
Because it doesn't make sense to zero-rate niche sites. By the same vein I could say Wikipedia competes with a knitting site run by my friend because if you want to know about knitting, I'd go to a knitting site, and it's not zero rated. That doesn't make much sense to me - if you want to enable minimal service to people, you choose something with broadest coverage, not with deepest one, because deepest one would necessarily be niche.
Even Mathworld/Weisstein's World of Mathematics isn't always right either.
I found an error once, and sent it in (it was something to do with graph theory, and the equation was missing a +1 or something, which was sort of critical). I checked about a month later and it hadn't been fixed, I can't remember what it was now, I really wish I could. It was something to do with "n/2 + 1" and they had written "n/2".
So how long till someone makes a VPN that uses Wikipedia edits as the underlying communication channel? The latency would be on the order of seconds but the throughput would be decent.
To those of you who think I'm joking, it's already been done with DNS lookups so why not here too?
Well in Angola they only have access to Wikipedia and Facebook, so Angolan users started to hide movies in JPEGs and PDFS, and link them in private Facebook Groups. [0] The issue with using edits for a VPN is that it will likely require a fast edit rate which would automatically get you flagged as a bot account.
Meanwhile in People's Republic of China, a peaceful (as in not troubled by war) country, People's access to Wikipedia still depends on the mood of those in power...
I would much rather see devices fitted with full dumps of the text content of Wikipedia(s). The english Wikipedia is just 60 Gigabytes for heaven's sake!
Most smartphones in Iraq probably have 4–16gb of storage, and 64gb flash storage would be prohibitively expensive for most Iraqis. The English Wikipedia is 60gb, but that's without _any_ multimedia. The total size of all media in Wikimedia Commons is 23 terabytes.
We could create a minified version of the Iraq Wikipedia that could be around 40gb, but many will not want it, so we'd have to sell Wikipedia phones and non-Wikipedia phones.
You could sell copies of the minified Wikipedia on microSD cards, but a 64gb microSD card still costs about $20, but that price is out of reach for most in the war torn country.
Then there's the issue of updated content. One of the main uses for Wikipedia is learning about current events, current status of crises, recent elections, recent deaths, etc. You wouldn't really be able to do that with this system.
Text only compressed is even less, in the 10-15GB range. I used to have it on an SD card and used https://www.rockbox.org/tracker/task/4755 to put in on a cheap media player.
https://en.wikipedia.org/wiki/WikiReader was one such attempt. Though reading the limitations, it seemed to do some heavy processing that threw away data to save space.
Well, I don't know anything at all about wikimedia infrastructure, but I would think it caches rather well.
Then all the ISP would have to do is coordinate with wikimedia to put such a device in their datacenter and save a lot on that precious upstream bandwith, keeping traffic on their local network where it's not costing them as much.
I would imagine that on average a page is retrieved more than it's edited. So even even in the case of poplar pages, there is potential for saving bandwidth.
But I don't know the actual statistics for wikipedia.
For Wikipedia, if you aren't logged in (with an editor's account etc.) you are hitting a cache most of the time.
And it isn't so much for the bandwidth saving, more the saving in CPU time in rendering the page up from the Wikitext source, which can be substantial. Some very big pages used to take 2-3 minutes to render because Wikitext is a rather slow ugly programming language pretending to be a document format. (A lot has been sped up after a lot of the rendering guts got moved to Lua.)
> Wikimedia recognizes the user is on that operator's network and serves a banner on the top of the page indicating free data courtesy of their mobile operator, which reinforces a positive brand experience for the operator.
I would like to know just how different Wikipedia Zero is. Can people still make accounts and contribute articles with this service?
> Through the Wikipedia Zero program, mobile data fees are waived for subscribers of participating mobile operators so that they may read and edit Wikipedia without using any of their mobile data.
To sidestep any straw-man arguments about war zones, different country, etc; What if this were implemented in Appalachia to give access to wikipedia to the poor an uneducated there?