6 ms·

Congrats on taking on mapmyride -- I used mapmywalk recently and their "new" UI is quite rubbish -- I use GMaps now but it certainly lacks some features. The d

Congrats on taking on mapmyride -- I used mapmywalk recently and their "new" UI is quite rubbish -- I use GMaps now but it certainly lacks some features. The desktop concept seems like a lovely idea-- but just be careful about getting too caught up in it. Desktop software (cross-platform?) can be a serious time sink. Make a better product and you won't need to import their data. For the users who are begging for it maybe try to hack together some desktop software with a scripting language -- don't worry about making it too polished, just make it work with as minimal as work as possible. Stay focused on the core user experience and you'll find plenty of users regardless. BTW, you can take some pleasure in knowing they wouldn't be calling in the lawyers if they really had confidence in their product.

cullenking 16y ago
I have no intentions of this being a nice, well supported and regularly maintained piece of software. It would be a single days work to hack together a ruby/nokogiri page scraping client and a simple UI. Application phones home to get new regexes or whatnot in the case that the UI changes on the pages to be scraped. But, your point is well taken. I am fond of SaaS rather than installable software for a reason!
ScottWhigham 16y ago
"BTW, you can take some pleasure in knowing they wouldn't be calling in the lawyers if they really had confidence in their product." I'm sorry but that isn't remotely close to real life, is it? Look - they are calling in the lawyers because a competitor is making itself more attractive to mapmyride's customers by using mapmyrides' services against their TOS. They are smart to call in the lawyers - it would be foolish not to. Also, who can say when/whether OP's company could "accidently" submit too many requests too quickly thus incapacitating mapmyride's servers?
lelele 16y ago
He doesn't need a desktop app. A Javascript app (thus client-side) would suffice.
- semanticist 16y ago
  Same origin policy would stop it from scraping MMR's site unless it was hosted there.
  - lelele 16y ago
    Can you enforce this? Here in Italy some websites forbid crawlers. I guess an user can always run a software over a website, especially on data entered by herself, can't she? Isn't a browser a scraper, too? Just make the user load a page and then harvest data from the DOM.
    - semanticist 16y ago
      The browser enforces it, not the website. You can't use in-browser JavaScript to load a page from a site other than the one you downloaded the JavaScript from. Otherwise any site you visited could load gmail in the background and read all your email. It's inconvenient when you're making client-side applications, but for the web to have any kind of security it's absolutely essential.
      - lelele 16y ago
        Thank you for your explanation. I've tried it. I'm a system programmer, not a web developer, and it shows ;-)
- cullenking 16y ago
  I am pretty sure that origin policies would prevent this from working. I considered using Adobe Air, but, it's probably easier to just bundle a ruby desktop app with a simple UI. nokogiri trumps all when it comes to screen scraping!

cullenking 16y ago

I have no intentions of this being a nice, well supported and regularly maintained piece of software. It would be a single days work to hack together a ruby/nokogiri page scraping client and a simple UI. Application phones home to get new regexes or whatnot in the case that the UI changes on the pages to be scraped. But, your point is well taken. I am fond of SaaS rather than installable software for a reason!

ScottWhigham 16y ago

"BTW, you can take some pleasure in knowing they wouldn't be calling in the lawyers if they really had confidence in their product." I'm sorry but that isn't remotely close to real life, is it? Look - they are calling in the lawyers because a competitor is making itself more attractive to mapmyride's customers by using mapmyrides' services against their TOS. They are smart to call in the lawyers - it would be foolish not to. Also, who can say when/whether OP's company could "accidently" submit too many requests too quickly thus incapacitating mapmyride's servers?

lelele 16y ago

He doesn't need a desktop app. A Javascript app (thus client-side) would suffice.

semanticist 16y ago
Same origin policy would stop it from scraping MMR's site unless it was hosted there.
- lelele 16y ago
  Can you enforce this? Here in Italy some websites forbid crawlers. I guess an user can always run a software over a website, especially on data entered by herself, can't she? Isn't a browser a scraper, too? Just make the user load a page and then harvest data from the DOM.
  - semanticist 16y ago
    The browser enforces it, not the website. You can't use in-browser JavaScript to load a page from a site other than the one you downloaded the JavaScript from. Otherwise any site you visited could load gmail in the background and read all your email. It's inconvenient when you're making client-side applications, but for the web to have any kind of security it's absolutely essential.
    - lelele 16y ago
      Thank you for your explanation. I've tried it. I'm a system programmer, not a web developer, and it shows ;-)
cullenking 16y ago
I am pretty sure that origin policies would prevent this from working. I considered using Adobe Air, but, it's probably easier to just bundle a ruby desktop app with a simple UI. nokogiri trumps all when it comes to screen scraping!