Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Pulse — A simple pulsating indicator. (moeedm.com)
163 points by moeedm on Sept 18, 2011 | hide | past | favorite | 40 comments



Top of hacker news: a dot that.... pulses. And doesn't even work in most browsers.

I mean I know it's the weekend and maybe a slow news day, but really?


Stick it on Github as three files - image, html, css - so folks can "watch" it. I've found that the watch feature of github helps me remember all these little nifty snippets of functionality so that, one day, when the need comes around, I actually remember that the thing exists, I can find it easily, and I can use it.


The only problem about that on GitHub is that you can't "unwatch" all the notifications while still watching the project. I'm watching 128 repositories and my news feed is just noise to me now. :-/


GitHub needs a major overhaul of the entire activity stream. Right now I'm forced to use the GitMarks Chrome Extension and keep a very curated list of projects / people to watch. And the feed is still chaotic and unusable most of the time. They overhauled Issues from a useless issue tracker into my primary Project Management tool, so I'm quite confident that they can revamp the actions feed successfully.


Someone from GitHub needs to see this comment.


I follow my friend's hobby projects, but my news feed is 99% CoffeeScript updates. It'd be really nice to have more control over that view.


I do that, too.

And same complaints as one of this post childs.

Sometimes though, I wonder why I don't use the tools which were made for this. I.e. the browser bookmarks.

Maybe the browser bookmarks really need some improvements to be useable. Because in many cases, I tend not to use them (leave open tabs instead, use 'watch project' on GitHub, and so on).

E.g., there could be some rules, like when I bookmark a page on GitHub, it automatically puts it into a bookmark directory called "GitHub projects" or so.

Maybe also more dynamics. Like there could be scripts running on a bookmark directory which check for updates on the specific site/project and push them up, print a headline or so.


On it.


I do not get it. All I see is a blue dot. Clicking it says "target acquired" but I can not acquire a new target...


It only works on webkit browsers.


Using JavaScript to render to a canvas will work in all modern browsers, while maintaining (or even enhancing) the CSS's flexibility. I first thought the JavaScript solution would even be about the length of Moeed's CSS, but after writing it out my code's about 5x longer :(

edit: here's a jsfiddle of it in action http://jsfiddle.net/MftVp/


With jQuery I doubt it would be 5x longer.


jQuery doesn't provide anything that would make the above code shorter.


Firefox 5+ supports CSS3 animations and transforms with the -moz prefix. A quick replace all in pulse.css gets the animation working fine.


This effect can be realized on every JavaScript-supporting browser, or in fact any animated-.gif-supporting browser.

I think it was a proof-of-concept work.


I can't but I'd downvote this simply for not being cross browser compatible in some form


An idea can be interesting without being cross-browser. The author makes no claims about this being the defacto-enterprise-grade blinking dot.


Might be nice if the link title mentioned it, so save the rest of us the time.


I think the idea of a pulsating dot has been around long enough that it could have been implemented a bit more solid.


Are you using an image element to support printing? There's no other reasons to not use CSS3 to create the dot (or the somewhat more pragmatic approach of setting it as a background image on the target element, reducing your markup), as I see it?

I think you should also expect that people could want more than one target on their maps. There's no downside to switching from IDs to classes here.


I understand it is a very neat piece of CSS scripting, but how does this compare to a simple animated GIF from practical perspective?


GIF's don't support alpha transparency, making them the opposite of practical for an effect like this.


At least GIFs work more or less the same on all browsers. This is an excellent example of how not to do it.

Imagine the support call: "Click on the blinking dot" : "I don't see any blinking dot" : "It blinks for me, do you use a webkit browser?" : "A what?"


CSS beats graphics because it can be changed programatically. Browser supporting, _all_ the parameters can be changed to suit the rest of the UI. If the designer writes their CSS in browser like I do, it's instantaneous and immensely powerful.

With images, it's, well, you know, slower.


Well you could script it to change the rate of pulsing in JS I guess... Still don't think this is very exciting anyways. Why the png and not pure a pure css icon for instance (although that's not very interesting to me either)?


I recently did something similar for more general elements as a jquery plugin: http://jquerypulsate.kilianvalkhof.com/ it uses the outline-offset css property, which doesn't seem that well known but seems quite useful too.


Very nice, I see myself using this before friday :)


Nice little effect. Could work well on realtime force-directed graphs:

http://mbostock.github.com/d3/ex/force.html


Last year I (along with the rest of our Node Knockout team) built a game that had a pulsing dot similar to this one. We used Raphael (SVG/VML). Although it's much heavier than this using Raphael is cross-browser. It might be a suitable fallback.

I want to open source the whole project but haven't had time to clean it up.


My suggestion: don't bother cleaning it up, just publish it. If the two realistic options are your code dropping off the face of the earth because it's never published, or it being put in a place where someone might find it useful and run with it, the latter is far more beneficial to society.

If you're concerned about your personal reputation (which in my experience tends to be the case) then consider you're likely better off as a prolific coder than a sparse one, and if anything we should be pushing the industry towards more open exchange of code so as to reduce loss & waste.


By clean it up I mean make it work again. It was built on an older version of Node and doesn't work with API changes made since then.

What you say is true though. I'll set a deadline to fix it by. If it isn't done by then I'll just put it up.

My code is always perfect in every way so I'm not worried about reputation. /sarcasm


That, plus the fact that if people like it they'll fork and clean it for you. FOSS + karma ftw.


Pretty nice, but it appears that whatever method is used to generate the pulse, it causes the CPU utilization of the Desktop Windows Manager (dwm.exe) to shoot up to about 10-15% when using Chrome.


<blink> 2.0?


I must agree with most people complaining here, it should work on ALL modern browsers (including IE9).

The implementation is not even great either, see this screenr I made: http://screenr.com/azAs


I'll be more impressed when this can work on my non-JS enabled browser.


doesn't pulse on chrome 14.0.835.165m on windows


I would make the pulsing ring easier to see


haha this looks like the one on Foursquare, neat work regardless though :)


A little bit (i.e. completely) derivative of the Maps app on the iPhone.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: