3 ms·

Every software shop in the world should be designing and coding for low connectivity speeds.

• johansch 11y ago
  I/they (I am no longer with Opera Software) did that for a decade with Opera Mini. It's an interesting engineering challenge. It got us lots of users (0.25B monthly actives) but they are (relatively speaking) quite poor, so ad revenues aren't as high as you might wish. So you need a lean operation, at least if your operating cost is significant - like transcoding web pages for 1/30th of the global population...
• elros 11y ago
  Why? Edit: It seems to me that depending on the market one targets there's not much to be gained from designing for slow connection speeds...
  • mynameisvlad 11y ago
    Probably because you don't know where your app will blow up. Unless the app is geographically restricted, there's no real telling where the spark will happen. Maybe it unintentionally solves a massive problem that is happening right now in India, and you suddenly have a massive influx of 2G users. If you hadn't even considered the experience for them, it won't matter how quickly it blows up; it'll fizzle immediately when you have a shitty experience. For example, on my trip a few weeks ago, I used T-Mobile's free 2G roaming to get around. Searching for something on Google was pretty quick, and loading a Wikipedia page also was relatively fast. But the second I moved away from them and onto real sites that didn't try to optimize for lower speeds, I was pulling my hair out at how much data was trying to get sucked down on such a slow connection. It took minutes to load some websites.
  • JustSomeNobody 11y ago
    Happy visitors to your page/app/whatever? Return visitors to your page/app/whatever? Just because one is connected using Verizon, doesn't mean the connection speeds at any given time are all that great.
  • izacus 11y ago
    Except that you get slow connection speeds pretty much everywhere as soon as you move outside your house. Subways, commutes, etc. For crying out loud, you even go to EDGE on AT&T when commuting in Sillicon Valley on CalTrain!
    • azar1 11y ago
      Exactly. This article really changed the way I thought about optimizing for slow network speeds: http://www.kryogenix.org/code/browser/why-availability/ http://www.kryogenix.org/code/browser/why-availability/
  • xixi77 11y ago
    The question is, how do you know how much is to be gained, unless you try? Just because you see most customers being on high speed connections most of the time doesn't mean much: for one, it could be because otherwise it's so bad no one does it; plus, a better experience for slow connections/older devices is quite likely to translate to better experience for everyone. Otherwise, you might figure it out when you suddenly lose most of your users to a competitor who offers a snappier experience.
• tsunamifury 11y ago
  No every software shop in the world should be designing to the requirements of their target userbase -- which often is not India.
  • bhhaskin 11y ago
    This is the correct answer. A parallel is also designing for outdated internet browsers.
  • kamaal 11y ago
    >>which often is not India. But its a often a low end phone with slow internet speeds. If I remember correctly, one of Google's advantage over their competition in their early days was their page would load quickly on low end desktops and slow dial up connections.
• kamaal 11y ago
  I read a quote a while back. If you want to develop fast software, use a slow computer. Not only should app developers use slower internet speeds, they should be developing it for the low end smart phones- Which make up for most of people who are using smart phones. Especially in a country like India, there are a lot of people using $20-$30 smartphones. Who make most of the users out there. Develop for them, and you get a huge user base for free.
  • toephu2 11y ago
    You get what you pay for... for many U.S. internet/tech companies India is typically at the bottom of the list of countries where the company's revenue is derived from.