|
fuf posted:omg Internet Explorer 8 fuf posted:2) What are people's thoughts on displaying a "You are using an outdated browser, you should update" message for IE8 users? Do you think maybe it comes across badly and annoys users? fuf posted:3) Bootstrap etc advocate a "mobile first" approach where you default to a one-column mobile layout and then use media queries to adjust for wider screens. But IE8 doesn't support media queries. So do you a) write a whole separate stylesheet for IE8? b) use a polyfill to get media queries working in IE8, or c) forget "mobile first" and make your default layout the one that IE8 users will see?
|
# ? Apr 6, 2014 17:00 |
|
|
# ? May 9, 2024 15:05 |
|
fuf posted:omg Internet Explorer 8 Remember that a non-trivial percentage if people using IE 8 simply cannot update or use anything else due to organizationa policy. As for #3, if your site is usable even if you delete your media queries, who cares if IE8 gets the "mobile" version. That said, if your mobile container rule isn't width:100%;max-width:some-sane-number-based-on-your-type-size then you should switch frameworks or tweak yours so it is that way.
|
# ? Apr 6, 2014 17:04 |
|
We dropped IE8 from our default testing stack a few months ago, and life has been so much better ever since. If clients want/need IE8 support, they can pay extra for it. That is all I have to contribute here except to agree that you should look at stats for your own audience before taking our advice.
|
# ? Apr 6, 2014 17:13 |
|
Lumpy posted:Remember that a non-trivial percentage if people using IE 8 simply cannot update or use anything else due to organizationa policy. The only reason my firm updated away from IE8 is that our biggest/most important vendor put a "welp sorry dudes better upgrade" message up, and as a result we upgraded within a week after a VIP freaked out.
|
# ? Apr 6, 2014 17:18 |
|
The company I'm presently at loves to outsource operations so what happens is the majority of those sites require > IE8 and thus users need to use Chrome, but all the horrendous internal sites only work IE6-8 so the user has to constantly juggle back and forth. Still amazed that VMware vCloud Director has really poo poo browser support, who writes something new so broken? It doesn't work in IE8 or anything using Webkit, Firefox is the only reliable client.
|
# ? Apr 6, 2014 17:39 |
|
EAT THE EGGS RICOLA posted:The only reason my firm updated away from IE8 is that our biggest/most important vendor put a "welp sorry dudes better upgrade" message up, and as a result we upgraded within a week after a VIP freaked out. While that may indeed happen, I suspect your firm is the exception. My main point was that many organizations do not allow users to install or update software themselves.
|
# ? Apr 6, 2014 19:14 |
|
Speaking of Django and REST, I'm planning on using Highstock to recreate my old stock charting web app. I wanted to add more functionality to the old app and also make it more dependent on client-side scripting. I'm wondering how I should serve historical stock data to the script. In a prior attempt at reviving my app, I made a really rudimentary API that served an array of datetime-price data for a given stock symbol, meaning that it was generated server side. What I did was use Pandas' Yahoo! Finance API functionality to get all the stock data, processed it, and served a JS array when a stock symbol was queried. Also, Django-REST-Framework seems focused on serving data from a database, rather than helping me serve something like that. I don't know if it will help me in this case. I can't help but think that processing on and getting the data from the server is not very efficient, but I don't know what else I could use. Yahoo! used to serve historical data through YQL, but the historicaldata table has been blocked. I could also get the CSV of the data like so, but I don't know if I'd get blocked from doing a bunch of queries on that. Is making an API solely to serve historical stock data still really that bad an idea? Pollyanna fucked around with this message at 04:36 on Apr 7, 2014 |
# ? Apr 7, 2014 04:33 |
|
I'm looking for ideas/inspiration for a modern-looking way for adding a related object in a CRUD-style app. For example, say I have a form for adding a Job to the users database. A Job can have associated Vendors and the form allows the user to select an existing Vendor or add a new one. What's a nice way of doing this? My current dumb version has a section of the form that has a list of currently added vendors, a section for selecting an existing vendor, and a link that pops up a modal with a form for adding a new vendor. Any other ideas, or can you point me to any sites doing this in an interesting way?
|
# ? Apr 7, 2014 21:55 |
|
Thermopyle posted:I'm looking for ideas/inspiration for a modern-looking way for adding a related object in a CRUD-style app. This is default behavior for has_many/belongs_to/children relationships in Hobo, which is a superfast prototyping thing for Rails. I would not really recommend it for anything serious at all, but I've used it a bunch of times when I've tried to get a fully functional app up and running in a couple hours.
|
# ? Apr 7, 2014 22:06 |
|
Thermopyle posted:I'm looking for ideas/inspiration for a modern-looking way for adding a related object in a CRUD-style app. Are you talking about the UI? If so, you could put a "New" option in the select list that shows the form inline on select.
|
# ? Apr 8, 2014 02:16 |
|
Kobayashi posted:Are you talking about the UI? Yes.
|
# ? Apr 8, 2014 02:41 |
|
EAT THE EGGS RICOLA posted:This is default behavior for has_many/belongs_to/children relationships in Hobo, which is a superfast prototyping thing for Rails. I would not really recommend it for anything serious at all, but I've used it a bunch of times when I've tried to get a fully functional app up and running in a couple hours. Wow, Hobo 2.0's screencast intro looks really cool. Definitely made me want to spend a day or two messing around with it. What are some of your concerns with it in production?
|
# ? Apr 8, 2014 11:16 |
|
http://heartbleed.com/ Haven't seen this posted in the last few pages. We have a shiny new openSSL vulnerability please update your servers people. You can check your domain here: http://filippo.io/Heartbleed ps. I won my contract negotiation and the non-compete was removed. edit: added some clarity. changed the test site, no more little checkbox. Oh My Science fucked around with this message at 21:04 on Apr 8, 2014 |
# ? Apr 8, 2014 19:54 |
^^^ Be sure to tick the "Do not show the results on the boards" on the above test Good to hear about the contract Oh My Science!
|
|
# ? Apr 8, 2014 19:57 |
|
Oh My Science posted:ps. I won my contract negotiation and the non-compete was removed. Contgrats! How'd you eventually get them to give in?
|
# ? Apr 9, 2014 14:56 |
kedo posted:Contgrats! How'd you eventually get them to give in? "Let me ask you a question -- do you really want to go to all this effort to hire a guy stupid enough to sign a contract with those clauses in it?"
|
|
# ? Apr 9, 2014 15:06 |
|
Thermopyle posted:Yes. I'm building something similar for my work at the minute, using basically the same visual pattern as a todo list. So you'd concatenate your separate lists + modal into one block (delineated by colour/horizontal separator/whatever), with the final list entry an 'add new' button. Expand/contract to display/enter further info, rather than pulling in a modal. Dead simple, basically stripping form styling and rebuilding with CSS/JS. I had some nice examples which I'll try and dig out. You get a very fluid interface (with simpler, more structured styling in many ways), and better ui performance on smaller viewports (target areas more obviously delineated for one thing, & my test users liked the flow much better than obvious form elements). Forces you to think about it (from structural/design side) as a more reactive interface, rather than as a bog standard form, which is a good thing, plus the interaction/transition methods and structures start to write themselves pretty quickly as well.
|
# ? Apr 9, 2014 16:06 |
|
RobertKerans posted:I'm building something similar for my work at the minute, using basically the same visual pattern as a todo list. So you'd concatenate your separate lists + modal into one block (delineated by colour/horizontal separator/whatever), with the final list entry an 'add new' button. Expand/contract to display/enter further info, rather than pulling in a modal. Dead simple, basically stripping form styling and rebuilding with CSS/JS. I had some nice examples which I'll try and dig out. If you have a lot of entries, making 'add new' the last one in a list can make discovery of that option very hard. Make sure you test with 100 or 500 entries.
|
# ? Apr 9, 2014 16:33 |
|
kedo posted:Contgrats! How'd you eventually get them to give in? I consulted a laywer and gave them the option of adding a consideration clause or dropping the non-compete, I'm guessing my bosses lawyer didn't laugh at that. They didn't want to give me a years worth of salary when I leave. please check your servers for heartbleed vulnerability Sorry, it's worth repeating.
|
# ? Apr 9, 2014 17:14 |
|
fuf posted:omg Internet Explorer 8 This is me. Right now. If we didn't have 13% of our userbase alone with the client using IE8, I would have drop-kicked IE8 support to begin with. Alas.
|
# ? Apr 9, 2014 18:25 |
|
Lumpy posted:If you have a lot of entries, making 'add new' the last one in a list can make discovery of that option very hard. Make sure you test with 100 or 500 entries. Yeah, we've the advantage that there's a (low) fixed upper limit. With large amounts, I'd probably look at using a search field with parameters to allow users to rapidly drill down through whatever list/select solution was used.
|
# ? Apr 10, 2014 06:15 |
|
Potentially dumb question I'm having a hard time googling for whatever reason: What's an easy way to set something on a parent container and all child containers that tells the browser "render this when you try to print, but not anything above this in the DOM" ?
|
# ? Apr 10, 2014 15:26 |
What exactly are you trying to do? I suppose you could add a css entry for everything NOT of the ID/class of the container and set display:none, but that's super hacky - :not doesn't really work well in IE I don't think. You can't change the html layout and just throw everything prior to that in a container that then gets hidden..?
|
|
# ? Apr 10, 2014 15:32 |
|
2banks1swap.avi posted:Potentially dumb question I'm having a hard time googling for whatever reason: code:
code:
|
# ? Apr 10, 2014 17:18 |
|
Couldn't you just do something like:code:
|
# ? Apr 10, 2014 17:39 |
|
Subjunctive posted:
The only problem with this is that visibility: collapse; only works on table rows and table row groups (thead, etc.) so this may not do what you intend.
|
# ? Apr 10, 2014 17:43 |
|
cbirdsong posted:Couldn't you just do something like: It would still show the elements in the parent chain (their borders, text nodes, etc.) and they would affect positioning, even if you spelled out the full descendent hierarchy. (In which case please x-post to Coding Horrors.)
|
# ? Apr 10, 2014 17:44 |
|
Edit: To be less of a chode I'll actually answer. Sulla-Marius 88 posted:What exactly are you trying to do? I suppose you could add a css entry for everything NOT of the ID/class of the container and set display:none, but that's super hacky - :not doesn't really work well in IE I don't think. Our project uses a top nav bar, left hand nav bar, and even a tab control bar. All the actually "stuff" anyone would ever want to print is inside of a content pane. I want to make that happen. I want to let that content pane be printed/rendered when it's print time, but keep everything else from being rendered. Also, ideally, if the content pane is scrollable (there's stuff stretching vertically) then I'd like it to render as such and print multiple pages. Fuck them fucked around with this message at 20:50 on Apr 10, 2014 |
# ? Apr 10, 2014 19:44 |
|
2banks1swap.avi posted:Edit: To be less of a chode I'll actually answer. Add a .noprint class to your top nav bar, left hand nav bar, and tab control bar. Your print stylesheet should have this rule: .noprint { display: none; }. Hooray!
|
# ? Apr 10, 2014 21:21 |
|
Got a problem with iframes. If you go to any product on this site and add it to your cart, you'll get a pop-up iframe quick cart. On the cart is a section for "Customers who bought these items, also bought:", and if you click any link in that section, it'll take you to that product. The problem is, it stays in the iframe, which is undesirable. I've tried putting target="_parent" on the links, as well as target="_top", but neither option fixed it. Halp? Should I just not bother working with their iframe?
|
# ? Apr 11, 2014 03:14 |
|
I use the Zurb email inliner for email markup. They always output the <style> tags in the body of the HTML. Isn't that incorrect? I always move it back up into the <head> section but I was just curious why they'd do it specifically for emails.
|
# ? Apr 11, 2014 19:38 |
|
Some email clients strip out the <head> entirely, so I'd imagine that's why they move it to the body.
|
# ? Apr 11, 2014 20:34 |
|
Yeah. Always have the styling applied as close as possible it's content, lest it get stripped out or misapplied. HTML email
|
# ? Apr 11, 2014 20:38 |
|
kedo posted:Some email clients strip out the <head> entirely, so I'd imagine that's why they move it to the body. And good email builders ensure all styles are inline!
|
# ? Apr 11, 2014 21:30 |
|
Widely-compatible HTML email is a dark art. Recommend not messing with the ritual or its components. Edit: really, it's best to just not treat it as related to web pages at all. Think of it as RTF with angle brackets or something, because lots of page semantics aren't honoured (internal links, resource references, sometimes the flow calculations). Subjunctive fucked around with this message at 22:34 on Apr 11, 2014 |
# ? Apr 11, 2014 22:31 |
|
Subjunctive posted:Widely-compatible HTML email is a dark art. Recommend not messing with the ritual or its components. OK, we can fix this guys! All we need to do is create a whole bunch of pages / reports that show that plain-text email has a higher click-through rate than HTML email. Then we pass each other our links, get them bandied about on twitter and blogs and so on, then show our bosses our "research" on how much time and money the company will save for better results buy not doing HTML emails any more. Foolproof I say!
|
# ? Apr 11, 2014 22:34 |
|
Just go the Apple route and have every single thing inside an image. It's like early 2000s web development all over again! Use images for everything! Need a headline in a font other than Arial or Verdana? Text in an image! Need to push that table ten pixels lower on the page? Spacer.gif! Need to give people a catchy way to get in touch with you? e: Real solution – offer "branded" MailChimp templates. Charge a ridiculous hourly rate to "brand" MailChimp's existing, free templates. Branding involves changing colors in CSS and nothing else. Easiest money I make by far.
|
# ? Apr 11, 2014 22:38 |
|
Stealing this for my portfolio, thanks!
|
# ? Apr 11, 2014 22:53 |
|
Heskie posted:Stealing this for my portfolio, thanks! Remember, if you don't have at least 10 animated GIFs on your page, you won't get the user's attention.
|
# ? Apr 11, 2014 23:05 |
|
|
# ? May 9, 2024 15:05 |
|
It really sucks that email clients don't play Flash® intros.
|
# ? Apr 12, 2014 00:03 |