A petition to stop the usage of lightboxes
rohanprabhu.com
rohanprabhu.com
Lightboxes are great for showing zoomed images or image galleries. They're not welcome for spammy widgets, signup offers or ads. But then, what format do you prefer for spammy widgets or ads? None obviously, so lightboxes are not the problem.
I, for one, much prefer the latter. You can continue with reading the article, you can use all the window management you already know and you don't have to suffer the horrible effects that author of the lightbox library though you can't live without. Not to mention that on netbooks, because lightboxes are always contained within the size of your browser viewport, the "zoomed" content is often smaller than miniature on the site.
It's exactly the same as with nojs users—if they can block, they can unblock it. Especially if they expect a pop-up to appear, because they've just clicked on a miniature.
Thankfully, it seems that makers of security and antivirus software are changing their products to not restrict JavaScript and browser functions.
It's an arms race. Advertisers will always use the most in-your-face tool available. First it was banner ads, before Javascript and pop-ups became common. Banner blockers appeared. Then it was pop-ups, until the broader community and browser vendors decided that the S/N ratio for pop-ups was so low that we were better off killing them entirely. Then it was Flash, with the added benefit of literal noise... and now Flash-blocking is an essential component of any browser. The current advertising tool of choice is lightboxes and other floating divs: harder to automatically distinguish from actual desired content within HTML, but we're starting to see browser addons and tools that can do that blocking job too.
Currently gaining in popularity is requiring that an advertisement appear and be experienced in its entirety before any access to content at all. Hulu led the way on this. Well, unless you count television and radio which discovered that format about sixty or ninety years ago. The market response was the VCR then DVR once the technology was feasible. Next move: a "web DVR" that will experience the ads for you and "record" the web site until it can be played back without the interruption?
Built properly, a lightbox shouldn't get in the way of your desired behavior.
Who benefits from that option existing?
It's similar to the old problems with modal dialogs in desktop applications. Basically: don't use a modal dialog unless your situation absolutely requires it.
In general, modal dialogs are suitable for operations where it helps to have contextual information in the background, but you don't need to interact with it. A 'Save' dialog is a common example.
This thread on UX exchange has some more thoughts and examples:
http://uxexchange.com/questions/1153/when-should-you-use-a-m...
http://viewtext.org/article?url=http://rohanprabhu.com/?p=16...
...looks like the site is struggling to serve requests right now
http://rohanprabhu.com.nyud.net/?p=161
It preserves images better than ViewText does.
If OP knew his hosting wasn't very well suited for handling load spikes HN users can generate, he should have posted the original link behind the cdn.
At this point the only thing OP could do is to create a rule to redirect to .nyud.net link if the referrer equals HN.
At this point there might be nothing the OP can do except wait until the initial hump of traffic has died down: ifthe server really is completely bogged down he may not even be able to open SSH session to make any config changes.
For an article like this, that does not need any dynamic content, would be better served by an event based server as these tend to cope better on low-end boxes in the presence high demand for static content than thread or process based servers.
Or, like you say, stick with the current web server setup and post links to high volume sites via a cache like CoralCDN.
I actually did some testing at that moment, before I edited the part you have quoted (I added this later), and I'm confident OP could have probably rescued it if he wanted to.
But I agree, it's pretty much game over because you can't edit the link after you've submitted it. So if you're not prepared to handle dense traffic, it's a good idea to submit coralled link instead or put your website behind CloudFront temporarily.