stackexchange / stack-blog Goto Github PK
View Code? Open in Web Editor NEWStack Overflow Blog
Home Page: https://stackoverflow.blog/
License: MIT License
Stack Overflow Blog
Home Page: https://stackoverflow.blog/
License: MIT License
Two minor bugs:
Based on issues related to #18
Native pagination has two problems:
JS pagination has a major problem:
What I'm going to try doing is making AJAX based pagination by creating statically generated JSON with the right authors and posts information, to be lazy loaded. I think it's the right approach that will be a middle ground.
The comments displayed at
http://blog.stackexchange.com/2015/07/the-new-stack-exchange-blog/?cb=1
and at
http://blog.stackexchange.com/2015/07/the-new-stack-exchange-blog/
are different, even though the post is the same one. It appears that Disqus is treating the URL parameter cb=1 as something that distinguishes between posts.
(The parameter cb=1 is appended by the Community Bulletin on SE sites)
You should be able to specify a future date for when the post will go live. This would deprecate the "drafts" mode for any post.
Making sure this works on a hostname will make the transition easier. Right now there are a lot of references to the subdirectory /blog, but it's unclear how easy it will be to reconcile if we need to.
Started here #58 and did so for podcasts
but would be nice to have it restored for the other categories
There's a few outstanding issues that we need to cover, especially for home page, individual posts, contributors page
Steps to reproduce:
Debated whether or not to report this, since it could be useful on occasion for quickly paging as the page navigation links remain visible (unless there are uncached images that load after the page scrolls). But this seems like an odd use-case.
Right now the blog has a generic IIS 404 page:
http://blog.stackexchange.com/fakepage
We can use IIS custom error pages to serve a themed 404 page like the rest of our sites. Example:
We can also wire up Google Analytics to track the errors (Jason or Geoff can assist with that)
So far, I've used the same generic copy on all the channels on the top of the sidebar. We may want to update this to reflect the specific channel: the engineering channel has open source contributions there, a bit more focused on our devs, etc.
There are some integrations with community bulletin and /blogs. Spoke with @aalear about it.
Needs a way for people to publish posts that aren't visible on any of the channels
We're going to have to use the Disqus comment importer if we want to get the comments on the existing WordPress. This is something we'll have to try when there isn't a lot of activity on the existing blog.
Currently, all full page posts have lost their left-align, right-align, and original sizing. This is due to a catch all img
rule in style.less
as the default for all images and lack of support for the current import tools from WordPress. We want to restore the functionality. We can either:
Going to look into these options.
Steps to reproduce:
#page1
Also happens on author pages.
Probably also worth mentioning that trying to navigate directly to a page by editing the URL fragment (for instance, to #page45
) doesn't do anything unless you refresh the page after submitting.
If we're gonna be doing ajax loading for this, I'm thinking we need some popstate and hashchange handlers here.
I hope you guys didn't consciously choose to annoy mobile users… use this:
<meta name="viewport" content="initial-scale=1" />
Ref: http://dev.blog.stackexchange.com/authors/jeffatwood/#page47
Page navigation appears in the header:
...while the page count and current page indicator are in the footer:
Besides just being awkward having these two closely related UI elements positioned so far from each other, the latter is sorta... intruding. Should either put both in the header, or create space in/above the footer for them.
This was brought up by @tlimoncelli and a few others in the Diversity group about putting some Twitter content here. Could explore using http://www.twitterfeed.com/ for this sort of thing.
On an older post (Stack Overflow Trilogy) the images specified in markdown don't show up as can be seen in this screenshot:
On the same post the dates of comments seem to be off. All of those comments have the same date (Jun 29, 2015) .
We use Podtrac for analytics & such on the podcast. My understanding is that we do this by prepending http://www.podtrac.com/pts/redirect.mp3/
to the soundcloud file URL before pushing it out to the feed - so our feed URL actually looks like http://www.podtrac.com/pts/redirect.mp3/feeds.soundcloud.com/stream/XXXXXXXXX.mp3
.
I think it's that simple, but Alex Miller knows more.
From http://dev.blog.stackexchange.com/2015/06/stack-overflow-in-portuguese-now-with-less-beta/
This image has a caption that's rendering weirdly.
The sidebar that shows contributors doesn't match up with posts on that page and sometimes shows duplicate authors if they have consecutive posts.
http://dev.blog.stackexchange.com/tags/serverfault.com
Presumably because of the '.com' in the URL.
The client-side paged loads work for very small scale, but not for anything with a decent post count. Maintaining this would mean loading lots of resources on the client they'll very likely never see as well as create an ever-increasing initial page size with slower load.
Currently loading http://stackexchange.github.io/blog/ results in my browser loading 43.9MB of content over 1.6 minutes, including things like http://stackexchange.github.io/blog/images/category/podcasts.jpg (14.5MB), http://stackexchange.github.io/blog/images/StackExchange-12.jpg (11.7MB) and http://stackexchange.github.io/blog/images/StackExchange-19.jpg (11.6MB). Even though the majority of the bytes are currently made up of a few images, there are still hundreds of HTTP requests happening.
I think for this to be realistic on a client performance perspective, it'll need to use Jekyll's built-in pagination generation.
Related to this: the hero images need to be sized to something reasonable for a client to load, the http://stackexchange.github.io/blog/images/category/podcasts.jpg example is a 5560x3707 14.5MB image of which only a 667x150px window is seen for the hero image (to be safe on assumptions: we can double that for high-dpi to 1334x300).
Related to #18 - something that just needs to get done.
For myself, @ErinCorleyGray and @hairboat to work on. I've given both point people on the community and marketing teams a training on the workflow, it's time to reflect any gaps in the README.md
to make sure anyone knows how to use the blog. Some things to consider:
Anything else I'm missing?
New permalink scheme has broken links in the README
repro:
search for podcast
notice the result cc-wiki-dump « Blog – Stack Exchange
it links to http://blog.stackexchange.com/category/cc-wiki-dump/
which is an 404
I think it could/should link to this blog
The characters ❮
and ❯
are used for navigation but with the current fonts they don't show up on my machine (Windows 8.1, Chrome 43.0.2357.45) -- Hilariously they show up here as I'm editing this bug report in the <textarea>
but when I just view the bug report they don't show, so I'm not quite sure if we can simply fix it by changing the font or not.
One of the ways I used the WordPress blog was to find what changed in the early days of the site. For instance, I could find the blog post that described the changes to reputation payouts for questions and know more or less when that changed. But I also used it to find things the other way around. If I see something strange in the data at a particular time, I can often solve the mystery by searching for posts that month. The new blog doesn't seem to have that functionality. Would it be possible to implement some way to browse around a particular date?
I know Google will index the blog pretty quickly, but we should have a search box available on the site even if we just use Google's custom search.
Example, soundcloud link - though the soundcloud URLs are all screwed up.
I can't see the old Wordpress files anymore to make sure the players were in there to begin with, but I'm guessing they got lost somewhere in import.
On the existing blog, there's a feed: http://blog.stackoverflow.com/feed/
On the new blog, there's a corresponding feed: http://dev.blog.stackexchange.com/feed/
On the existing blog, there's a feed for podcasts: http://blog.stackoverflow.com/category/podcasts/feed/
On the new blog, there is not: http://dev.blog.stackexchange.com/category/podcasts/feed/
We get a fair number of hits on that feed. Apparently there are programs that allow people to subscribe to podcasts for the purpose of listening to them on mobile devices; anyway, we should try not to break that link.
Also, it would be handy to have feeds for tags in general.
It's confusing to have both, and we're more familiar with tags as a concept.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.