A website standard that is affordable to the poorest demographics in developing countries?

Fact: the Internet is excruciatingly slow in many developing countries, especially outside of the big cities.

Fact: today's websites are designed in such a way that they become practically impossible to navigate with connections in the order of, say, 512kps. Ram below 4GB and a 7-year old CPU are also a guarantee of a terrible experience.

Fact: operating systems are usually designed in such an obsolescence-inducing way as well.

Fact: the Internet is a massive source of free-flowing information and a medium of fast, cheap communication and networking.

Conclusion: lots of humans in the developing world are missing out on the benefits of a technology that could be amazingly empowering and enlightening.

I just came across this: what would the internet 2.0 have looked like in the 1980s. This threw me back to my first forays in Linux's command shell and how enamoured I became with its responsiveness and customizability. Back then my laptop had very little autonomy, and very few classrooms had plugs, but by switching to pure command mode I could spend the entire day at school taking notes (in LaTeX) without running out. But I switched back to the GUI environment as soon as I got the chance, because navigating the internet on the likes of Lynx is a pain in the neck.

As it turns out, I'm currently going through a course on energy distribution in isolated rural areas in developing countries. It's quite a fascinating topic, because of the very tight resource margins, the dramatic impact of societal considerations, and the need to tailor the technology to the existing natural renewable resources. And yet, there's actually a profit to be made investing in these projects; if managed properly, it's win-win.

And I was thinking that, after bringing them electricity and drinkable water, it might make sense to apply a similar cost-optimizing, shoestring-budget mentality to the Internet. We already have mobile apps and mobile web standards which are built with the mindset of "let's make this smartphone's battery last as long as possible".

Even then, (well-to-do, smartphone-buying) thrid-worlders are somewhat neglected: Samsung and the like have special chains of cheap Android smartphones for Africa and the Middle East. I used to own one; "this cool app that you want to try out is not available for use on this system" were a misery I had to get used to. 

It doesn't seem to be much of a stretch to do the same thing for outdated desktops. I've been in cybercafés in North Africa that still employ IBM Aptiva machines, mechanical keyboard and all—with a Linux operating system, though. Heck, I've seen town "pubs", way up in the hills, where the NES was still a big deal among the kids, not to mention old arcades—Guile's theme goes everywhere.

The logical thing to do would be to adapt a system that's less CPU intensive, mostly by toning down the graphics. A bare-bones, low-bandwith internet that would let kids worldwide read wikipedia, or classic literature, and even write fiction (by them, for them), that would let nationwide groups tweet to each other in real time, that would let people discuss projects and thoughts, converse and play, and do all of those amazing things you can do on the Internet, on a very, very tight budget, with very, very limited means. Internet is supposed to make knowledge and information free and universal. But there's an entry-level cost that most humans can't afford. I think we need to bridge that. What do you guys think?

 

 

Comments

sorted by
magical algorithm
Highlighting new comments since Today at 7:01 PM
Select new highlight date
All comments loaded

People are way ahead of you on this. For instance, Facebook Zero is Facebook's Potatophone-friendly site. It's text only and conserves data.

People are also trying to improve phone service and models in the developing world, and this has been pretty successful.

It's also structured in a way that facebook pays the mobile carries for the bandwidth.

I'm glad to hear that. But it'd be nice if the efforts were more widespread. (Weird that I can't use access it from my PC though).

Can you point to other concrete projects?

The Subsecretaria de Telecomunicaciones of Chile ruled that zero-rating services like Wikipedia Zero, Facebook Zero, and Google Free Zone, that subsidize mobile data usage, violate net neutrality laws and had to end the practice by June 1, 2014.

Wait, what?

David Meyer expresses a similar conviction in GigaOm, writing, “Broadly, though, it has the makings of a strategic disaster, stopping Facebook (and any other social network or messaging service) from entrenching itself in the mobile market in an unassailable way. And in the long term, for consumers, that is a very good thing indeed.”

Oh. Interesting.

The zero-rating mentioned is where the carriers don't charge customers for the data access to those services. This is commonly advertised in these countries along with the cell phone service ("Free Facebook!" pops up a lot in the Philippines, where people often sell sim cards on the street and many small general stores recharge cell phone plans -- adding some marginal pesos to your cell phone is often a pain). Pretty transparently not net neutrality, although if you are moving them from can't-afford-any-sites to can-only-afford-facebook, it's hard to see that as a bad thing, at least when you isolate it from the game theory / market capture elements, which are potent.

It's good to know that you care so much about people in the Third World. Congratulations on your moral excellence.

These problems frequently need to be tackled from multiple angles.

I recently heard about a company called Endaga who have come up with a single box that you just plug in to the power and an internet connection, and it creates a small-scale “cellular” i.e. mobile network, which then makes VoIP calls to the outside world. They’ve based it on OpenBTS, and they’re selling it at just $10k, with Endaga taking care of almost all configuration, interconnect, etc. They’re pitching this at rural areas in the third world where mobile coverage is sparse or non-existent. Obligatory link.

Note: I have no connection with Endaga, etc.

For some sites, text-based client apps may be a feasible alternative to Web browsers — but the same restrictions that sites use to deter spambots also make it hard to write text-based client programs that can post.

To find out how hard this would be, I just spent a little while investigating Twitter's APIs, which I've never used before.

A program to fetch and display a user's public Twitter posts is literally seven lines of Python; the hardest part of it is getting credentials from Twitter. Twitter requires that each client program have an "API secret" which is not supposed to be divulged to the user. Any given API secret is restricted in how many queries it can send, so if you share the same secret to everyone who uses your code base, it'll stop working. (Unless you pay Twitter, I suppose.)

This pretty much means that you can't hand everyone a Python script that lets them use Twitter from the command line; at minimum, each user would have to go on the Twitter API web site (logging in with a conventional browser) to get an API secret, then edit the script.

For posting or editing, Web APIs often require some sort of verification other than a username and password — such as a captcha answer, which depends on graphics. In the Twitter case, they won't let an account post via the API without an associated mobile phone number.

Whatever happens when the user logs into twitter with a conventional browser to get an api key is something you could do in your script.

Given how things in the third world look desktop doesn't seem to be the prime way to access the internet. Smartphones are much more promising.

If you don't have the bandwidth for images it's already possible to shut them down. You can use Wikipedia on low bandwidth.

As far as reading classical literature I doubt that the average child in North Africa has a desire to read classical literature. That child likely has a lot more pressing needs. It likely wouldn't even have the idea in the first place.

Internet is not only about technology but also a lot about culture. You can't simply look at what's nice for Westerners and then believe that the same thing will work in North Africa.

Using cell phone minutes as currency is for example an important feature in some countries with don't have good banking systems. In the West we don't use the technology that way.

It's possible that someone manages to program a Ripple based system that works better than trading cell phone minutes but that's not trival.

I don't think of this as "what would be nice for Westerners", I think of this as "what would have been nice for me, back when I was a kid, if I had known what I was missing out on". I may not know what the specific social norms or economic circusmtances of people from even a few hunderd km away from my hometown were (tribes in my country are insular like that), but I think I have a pretty firm grasp on what young nerds have always wanted throughout history and regardless of the culture they were born in, whenever they were given the chance.

(To give an example: If you took a bunch of normal people and gave them too much free time and wealth, and gathered them at a party, you woudn't get Plato's Symposium. You'd get the sort of lame party that's described in The Great Gatsby. Regardless of specific cultural or economical circumstances, It takes a special kind of mind to have a blast making stories up and speculating on the cosmos the way those folks had, and then to go and write that stuff down. And that special kind of mind pops up every where, and every time, as soon as you give them the chance to bloom. If you give them no resources whatsoever, they'll still go and, say, make up an epic poem or a creation myth or whatever. Minds that blaze will find something to burn, given even the slightest chance.)

We want to learn stuff, we want to read books, we want to talk about the books we read, we want to spin our own stories. We want to learn skills, build things, dream about distant lands and eras. We want to speculate about the meaning of life, and ask "why" at stuff until the whys run out.

We see the society we're in, the way people live, and some of it doesn't make sense, and some of it doesn't seem fair, and we want to do something about it, improve stuff, solve problems, or at least understand why things are the way they are. There's only so much we can get out of reading the Qur'an or asking our elders. Eventually we feel the need to ask further questions. And we want a space to discuss them with other people who care, other people like us.

Whenever EY references classic sci-fi literature, I feel envious because before I went to study in Europe I had only been able to get my hands on a couple of Asimov and Radbury books, despite how much I relished the stuff.

Plus, if we're the one clever boy in the village (heck, the one anything in the village), it's alienating, isolating. Why am I different? What's wrong with me? The Internet can help you find that you're not alone. That, no matter how unique your tastes or ideas or experiences, there are thousands, millions out there, who thought like you. This might be especially relieving to LGBT folks, and to people who aren't satisfied with their assigned roles in life (such as gender roles in sexist societies).

It also opens your opportunities. Again, the clever boy would become, say, the Witchdoctor's (or Curate's, or Imam's) apprentice, because that's what you were expected to do.

With the Internet, you don't need to move to the big city to get a good education, leaving your family alone, perhaps draining them dry. Your town school doesn't need to buy the books to make up a decent library. Heck, you may even not need to walk ten kilometers in the wilderness at the crack of dawn every morning just to go to the underpaid, ignorant schoolmaster's tiny classroom (who may or may not spare the rod).

But the Internet isn't just a path to bringing the Gospel of Western Civilization down to the poor isolated folksies. It's also a way for the folksies to communicate with each other, to gain visibility, to speak up for themselves, to tell their own stories in their own words, to document themselves.

The Internet is a tool of self-empowerment. And I believe it should be a human right, as much as clean water, warm homes, and lighting at night.

A Western kid has a lot more unscheduled free time than a kid in a town that lives on average on 1$ per day. It has already learnt reading and writing in school. The kid also likely can't speak English.

Clever as in two standard derivation higher IQ as the others in the village might also mean IQ of 110 in that village instead of IQ of 130. IQ might very well be a biased metric and there might be cultural reasons for scoring low on the test but people with lower IQ while have a harder time to figure out how to use a command line interface of a computer to subscribe to some mailing list on which they want to chat.

The Internet is a tool of self-empowerment. And I believe it should be a human right, as much as clean water, warm homes, and lighting at night.

That's a quite interesting claim. Part of what the debate about a human right to clean water is about, is whether it's the job of the government and the government should regulate it.

Cell phones companies in Africa don't face much regulation which allowed them to give people who don't have access to electricity because of overregulation access to cell phone coverage.

Not regulating the sector and simply let those companies provide more services and trust that Moore's law makes computers cheaper might be more promising than to start to declare internet a human right and let it be subject to heavy regulation.

I've been giving some thought to your "official rights to X result in governments taking ineffectual measures that result in people not getting X". I think it can be argued that if the measures stop people from getting X, they are vulnerating people's rights to get X in the first place, and are therefore "illegal". Does that make sense?

I also consulted my brother on this, who actually studies Law. His answer amounted to a very harried "It's a fracking mess. only experts can 'sort of' make sense of it."

... A thought just occurred to me. Just thinking out loud here. Do "fundamental rights" matter in the same way that religious dogma does? I mean, do they operate in a similar way, in terms of their tangible consequences on people's daily lives? Ecclesiastical Law. International Law. The Vatican. The UN. Dogma. Inarguable principles. Doctrine. Rampant hypocrisy and yet immense respect and authority. Saint vs. Merchant interactions.

Maybe I'm missing a vital difference, but the more I'm thinking about this, the more there seems to be a parallel of sorts...

Fundamental rights work, and are consistent, as long as one sticks to the so called negative rights or "first generation" rights, e.g., free speech, right to life, right to no arrest without a fair trial, etc.

For example a negative "right to internet access" simply means that the government can't bad you from using the internet, i.e., laws like this would be a violation. Neither it, nor anyone else, is actually obliged to provide you with a computer or any kind of ISP.

This is very close to what I work on full time at Google. I work on pagespeed which is open source, but the old [website mobilizer] (http://www.google.com/gwt/n) might be closer to what you're thinking of.

I'm happy to answer questions about these or about how workable various ideas are.

You could start a company that parsed webpages (maybe using an API like this one) and/or use a cache in order to automatically transform webpages to make them less bandwidth-heavy, then serve the webpages to people in the developing world.

Opera Turbo seems related but I have never used it and thus can't tell if it is actually any good.

Can you identify particular problem web sites?

I'm guessing that the major web sites, particularly the ones devoted to disseminating scientific or academic information, have already got this covered.

Is Wikipedia a problem? I doubt it. Google. Pubmed. xarchiv. I'd guess that Khan Academy is already on top of this issue as well.

Access with a tablet or a smartphone, and you're fine.

It doesn't seem to be much of a stretch to do the same thing for outdated desktops.

If those kinds of internet cafes are the particular issue you're concerned about, I'd think that changing the browsers to mimic a mobile device would go a long way toward solving bandwidth issues.

I'm guessing that the major web sites, particularly the ones devoted to disseminating scientific or academic information, have already got this covered.

Last time I was in Africa, google worked terribly, especially Gmail, Gdocs, and so on. You had to wait for a while before even being given the option of switching to the html version. And I had a legit wireless internet connection. 20$ a month, allegedly 2Mbps.

changing the browsers to mimic a mobile device would go a long way toward solving bandwidth issues.

Sounds like a great idea.

Oh, you're talking about using web apps and such. I'd expect latency would be the biggest issue. Even so, I'm surprised about the load time issue you saw. But they have to load the library for the app, don't they? (Built into chrome?) Maybe a chrome versioning issue? Update issue? And you were using chrome, right?

Might be better to install android/chrome on the desktop and use the app from there.

And better still to use a native android/chromeOS device.

I agree with Ritalin, and I suggest you try the internet in developing countries firsthand to get a taste of what it's like. It doesn't have anything to do with the browser version.

Experiencing poor internet speeds isn't high on my list of reasons to take a trip.

I use a Chromebook, man, and my phone is a Nexus. You can't get more chromey and googley than that.

Heck, sometimes the network gets crappy at my home, here in Spain, because of interference or because too many devices are anchored to the one router; even with a 50Mbps optic fibre connection, sometimes gmail will take five full minutes to load if I'm in my room. Networks can get slow anywhere.

I do believe that optimizing for low bandwidth and low CPU and RAM usage is universally useful. Slow-network problems can pop up anywhere. Plus, not everyone has the wealth to buy more than low-end machines.

It's just that the issue is exacerbated to the extreme in third-world countries, and one can't tell them "you should buy more expensive stuff" and "you should get more googley gear".

sometimes gmail will take five full minutes to load if I'm in my room.

On my 6 year old desktop, Gmail takes 5 seconds in chrome in my browser (which actually seemed kind of slow), and basically instantaneous on my nexus 5.

On my chrome book, it was a surprisingly horrible maybe 30 seconds on first load, and maybe 15 on subsequent loads. That is from the gmail link with chrome browser running, so it's bringing up the browser too.

So, gmail isn't the issue if you're waiting 5 minutes. Or maybe it's syncing all your email ever, and you have a lot more than I do? It's not my primary email.

The chrome web store has "gmail offline" which might be better for you? I find it annoying that it seems I always have to tell it whether or not to store email offline when I open it. But it opens quickly after that,

But it looks like the real trick for a chrome book is getting android apps to run. There are instructions on how to do this now. If you really need gmail, it might be worth it. In general, I think we'll have to just wait until Google works out android apps on chrome os.

But back to the overall point, I thought the issue was about information availability, and not really personal apps. Do you have a major information site that is a problem?

I'm just reporting my own problems the way they happen.

But back to the overall point, I thought the issue was about information availability, and not really personal apps. Do you have a major information site that is a problem?

Pretty much any site that employs graphic ads or that is otherwise graphic-heavy; wikias, news sites, fora... even dictionaries like thesaurus.com, wordreference, or leo.org can be a pain to load. The little menus that appear when one is making a search are especially sensitive to network lag. Thankfully there are offline, command-based equivalents to several of these dictionaries, but you have to look for them.

Plus, personal apps are extremely useful tools. To give just one simple example, why spend phone credit on calling or SMS when you can What's App? I'm sure we could come up with several app ideas that would have excellent time-saving or resource-saving capacities... and, once they learn how to code, who knows what the local whiz kids might come up with?

Are you running ad block? That might help with the graphic ad loads.

Again, I'd guess it's latency and not speed that's causing you problems with menus. Looks like interaction is the issue, not bulk data download. For general browsing, I don't know that you can do much about latency, but for particular apps. sites, if they've got an android app, you use it.

Seems like android apps are much better about all that stuff, and they'll be more generally available for chrome os soon, and I see some windows methods for this as well.

There are serious and well funded grownups working on this at Google.

Interesting to hear about the issues, though. You encouraged me to test out the chrome book I'm working on for my mother. Still slow by my standards, but I think it will be fine for her.