Show HN: Pagesnap – Take monthly screenshots of any webpage automatically | Hacker News Reader

Show HN: Pagesnap – Take monthly screenshots of any webpage automatically

pagesnap.io

62 pointszrail48 comments

I built something very similar last year and still reference it all the time:

https://www.dailysitesnap.com

There's a payment form but I never hooked anything up -- it's all free. You can sign up to have a screenshot of any URL emailed to you on a daily basis, or opt out of emails and just have it archived on the site. I've been collecting daily screenshots of ~20 public web sites for the last few months:

https://www.dailysitesnap.com/public

EDIT: fixed embarrassing typo, good morning everyone!

HN discussion from first iteration, SiteGazer:

https://news.ycombinator.com/item?id=6612286

But when I turned it into a real service, there was virtually no interest:

https://news.ycombinator.com/item?id=6654544

zrail, I've got some Ruby code that hits Selenium to take these screenshots. Email adam at cameralends if you're interested, you can have it too.

A few people emailed me, here's the meat of the code: https://gist.github.com/derwiki/9238362

Cool, thanks!

This was inspired by a short discussion on Twitter between mperham and patio11 last weekend.[1] The basic app works great right now but I haven't put the billing code in quite yet. If you're interested, sign up for the list[2] and I'll send you an invite (with a discount) when it's ready to go.

https://twitter.com/mperham/status/436976217443422208

http://www.pagesnap.io/list-signup

You are the Master of Modern Payments and the billing code is the only thing it lacks??

Just kidding! ;) Smart move and good luck!

Thanks!

I saw that same conversation and went looking around thinking there must be something for this. Cool to see it come to life so quickly, nice job.

Love the simple site and focus. I've been working on a similar project focusing more on visual diffing, but haven't been able to launch yet. Still building the server infrastructure, product focus, and working out the billing.

Thanks for sharing!

I did something like this for the 37signals homepage a while back. Used versions of the site from git:

http://signalvnoise.com/posts/3007-37signalscom-homepage-evo...

Along these lines, I've always wanted a repo specific wayback machine that would generate a state of your site for every commit, so you could browse what your site looked like at every point along the way as you built your repo.

So I could browse back to last week and see what my site looked like, or to my 3rd commit and see what my site looked like then.

I was thinking about that the other day, that would be awesome something that builds it at every step and converts it to an animated gif

what we need is some nielsen families that have full videos of their everyday surfing - just whatever they're doing, 24/7, saved for posterity. Imagine seeing how someone used the 'net in 1994 and 2004. It would be an interesting museum, to say the least.

Add dpxdt to your CI workflow.

https://github.com/bslatkin/dpxdt

  for commit in `git log --format="%H"`
  do
    git checkout $commit && build_my_site.sh && run_my_site.sh
    phantomjs screenshot.js $commit
  done

- Adding visual diffs would be very cool.

- Some sort of visual timeline would also be great, i.e. the ability to flip through screenshots with a javascript slider or something

- Export to gifs of video might be useful

- The ability to tag information to certain screenshots would be useful for noting changes and milestones (like in Google Analytics)

I recently started a project that takes screenshots of top news sites every hour:

http://newsabovethefold.com

It's really fascinating to look at once you've collected a decent number of screenshots to animate through:

http://newsabovethefold.com/animate/1 (CNN.com every hour for the last two months--will load slowly!)

I'd definitely be interested in using this service if I wasn't already paying for Snapito.

This is great! I love the mouse over zoom.

I actually had this idea a few years ago but never got around to building it, so I'm glad you're making it happen.

At my previous startup I hacked this manually by taking a screen grab with Evernote once a month.

I've heard Alexis Ohanian mention he is thankful for having the foresight to take screen shots or early reddit builds too.

I'm sure a lot of startups would find this useful for capturing the journey of the product and then later nostalgia.

Question. Is the number of urls such a huge burden?

At Theneeds.com we manage about 4k websites (planning to grow to 10k) so the prices of all these services are totally out of our budget, and we ended up with an in-house solution to take screenshots.

I would see myself paying more depending on the total number of screenshots and/or the size of the screenshots, but I really don't get the difference between 1 or 1k urls.

It's not necessarily a huge burden, but if someone wants to run 10K URLs through this thing on a daily basis we'd need to have a chat. As of today it's not ready to scale that big.

Of course, if someone did want to run that many through I'm sure we could talk about volume pricing, given sufficient lead time to scale the app.

Thanks, I'll follow up then :) (but give me some days as we just launched the iPhone app, we're all pretty busy)

Cool feature to add would be mobile and tablet versions

I'll add it to the list. Would customizing the width and user agent allow you to do what you want?

That would be a good idea and keep you from implementing all device requests!

Are the screenshots being grabbed by Pagesnap or the website's end? Do I have to include some sort of JS snippet? Could you use this to monitor the competition's landing page?

Screenshots are grabbed from the server. At the moment there's nothing you have to include on your end.

Excellent. Let me pay you.

http://www.pagesnap.io/list-signup

You can use this as well: http://visualping.io/. We're still working on doing more frequent checks on gowatchtower.com

@650REDHAIR .. try https://diphur.com if you're interested in monitoring competitors web pages.

I think the real benefit of sites like this are accountability and historic research. Having the ability to see what has been omitted or taken down is far more interesting.

Reading some old Steve Yegge posts (e.g. https://sites.google.com/site/steveyegge2/five-essential-pho...) I started wondering about the feasibility of a web service that would automatically change links on articles older than say, 5 years, to the equivalent Web Archive/Google cache/other links.

You can do this for broken links on your own server using code from the Internet Archive https://addons.mozilla.org/en-US/firefox/addon/resurrect-pag...

You can do this with a client-side extension, like Resurrect Pages, which presents links to the Google cache (for recent sites under load) and the Internet Archive (for old links that have rotted). https://addons.mozilla.org/en-US/firefox/addon/resurrect-pag...

Whenever I'm reading something on Reddit/HackerNews/etc that I think is going to be relevant later, I hit my Archive.org bookmarklet:

javascript:void(open('//web.archive.org/save/'+encodeURI(document.location)))

Which immediately saves the page to the Internet Archive. I haven't had time to look into if the Internet Archive has an API you can issue a URL to and it will fetch the content in the same way as above.

I do the same thing, except as of yet, I can find no way to search old site grabs in the wayback machine. AFAIK, the content is not indexed. You have to know it is there.

This is a good point. Added to my research list.

Wouldn't it just be javascript:void(open('//web.archive.org/web/*/encodeURI(document.location))) ?

Looks like it.

Very cool and I think the pricing is just right. Why are there 3 start now buttons all together though? Does each correspond to the plan you'd like to buy?

Thanks! Yeah, that's the idea. Sounds like the pricing table could use a little work.

That's a really good name.

http://archive.org/web/

Not daily but does the job

This would solve a large problem I have, if it could be taught how to click on a few well-named #ids my dynamic page has. It's not obvious whether it can do this.

Couldn't i use the free version that's somewhat more relevant: http://visualping.io ?

I love everything about this. Good job so far!

It even uses sidekiq under the hood! :)

but good job building it. Kudos to your creativity.

Reply on news.ycombinator.com