Still seeing the old Grist? New links won’t work? Here’s help!
Occasionally, when a site like ours moves locations on the web, a small portion of users gets a bit stranded, and can’t see the new version of the site. If this is happening to you, or if very recently published stories on Grist.org are not resolving for you, it’s very likely because your browser or ISP are improperly caching our old location through DNS.
Luckily, there are ways around this. Here they are:
- First, clear your local DNS cache. Browsers sometimes cache DNS for a long time … too long. Luckily, you can force them to stop. Here’s how.
- 99% of ISP’s are now replying with the correct address for Grist. If you’re in the unlucky 1%, you may wish to change your DNS servers for a while. These are often configurable in your internet settings. A good choice is to use Google’s public DNS servers, which tend to be reliable. We’ve tested these for Grist, and they work. Here’s how to switch on Windows, MacOS and Linux. Also, you may wish to complain to your ISP — what else on the internet are you missing!!??
We’re sorry that some of our readers have experienced this temporary problem. Please contact us if you need help getting access to Grist — we’re glad to help.
Grist is now on WordPress
Notice anything different on Grist.org today? I hope that the answer is “no”!
Even though you may not see any overt changes, today was relaunch day at Grist. We turned off our old site, and turned on a new, WordPress-based version.
What we’ve done is known as a “port” — that is, we’ve taken the Grist UX and content, jacked it up, and replaced the old foundations with new, better WordPress ones. This required many many weeks of effort, some valuable expert advice and lots and lots of coffee. It also meant that @natebot and I have had to learn the WordPress API rather quickly … luckily, this has been fun.
For more of a public intro to what we’ve done, please read this post on Grist by Scott Rosenberg, our Executive Editor. Notice a bug with the content or site? Please report it here.
Grist is adopting WordPress for two main reasons: WordPress has become the world’s foremost CMS, and is the locus for a whole lot of journalism innovation and experimentation. We want to be part of that community, and more able to take advantage of the latest online news technologies. Our other goal has more to do with who we are. We are small — really small — and as such want to focus our technical energies on news, design and innovation, rather than hosting, uptime and infrastructure. Our new hosting arrangement with WordPress.com VIP allows us to do that, and lets us join the ranks of other large media sites like Time, Chicago Tribune, and these guys from down the street.
So while today’s change might be almost invisible to most, we can tell you that there will be more exciting, and more rapid changes on the way, so stay tuned!
The News Smashup: 4 examples of a news app sub-genre
After attending one of Twitter’s developer teatimes here in Seattle and having various other Twitter API related experiences recently (more on that soon), it’s becoming a bit obvious that there’s a rapidly-emerging subspecies of news app coming into being. This sort of app can trace its origins pretty directly to pre-Twitter sites like Techmeme and its cousins, but can be distinguished by a primary reliance on social sources, rather than feeds (Techmeme etc. now also incorporates Twitter into its algorithm, but that was added later.)
That’s not to say there existed no news sites/apps based on the Twitter API before the last six months. However, whereas there is a lineage of Twitter mashups out there, these tend to be fairly rudimentary in nature in the sense that they don’t implement much (if any) intelligence on top of twitter streams, but simply aggregate and display them in clever ways. And while the big search engines and Twitter’s anointed resellers have been consuming the Twitter firehouse and implementing all sorts of intelligence/NLP/indexing over it for some time now, these sorts of activities (even on the level of link counting and indexing) were beyond the powers of news developers.
That has now changed.
So if we have a new genre on our hands, we have to name it right? I’m going to call these sorts of apps news smashups. (ok so it’s kind of a dumb name, and I know it can also mean “bad automobile accident” … but humor me.) Definition:
A news smashup is a news-oriented web application that:
- consumes streams of content primarily from social sources (Twitter) in near real time.
- applies intelligence/indexing/filtering to these sources, also in near real time.
- implements a web experience that to the end user appears to be edited, curated or guided by a subset of the social web.
Drawn completely from my own biases and experiences, here are a few illustrative (and in some cases mysterious) examples of this type of thing:
Muckrack! What started as everyone’s favorite Twitter mashup for collecting, categorizing and streaming tweets from a big pool of journalists is now replete with bells and whistles, including new pro features like search and alerts. This is a great example of a curation/aggregation Twitter application focused on a particular vertical. It’s also an example of how a mashup can become a … smashup!
From the somewhat mysterious department, there’s Percolate: a Twitter aggregator powered by a set of relevance algorithms and feedback mechanisms … but with much higher aspirations. From the horse’s mouth: Percolate works by hooking up to streams of content … and filtering down to the most interesting stuff for you . We then present that content back to you for you to react to, which is as easy as hitting an “awesome” button.
Sociative develops technology to use “social signals” (largely from Twitter) to identify content signal amid a huge amount of content noise. Applications of their technology so far include several socially-curated sites, most recently a socially driven representation of news from the occupy movement called Occupy Live.
Andrew Phelps, prototypical journo-hacker, built Fuego, The Nieman Lab’s “heat seeking twitter bot.” Fuego collects the most popular content from a curated set of media experts in realtime, and exposes them as a simple list of trending material based on link counting and other weighting.
I’m sure that this is an entirely incomplete list. I’d love to hear what others think about defining this as a genre. Does it make sense to do so? If so, what are some other examples of this sort of thing in the news apps space?
Infographic: Household Water Usage
Making several changes in your home can conserve water and save you money.
Water Infographic: All Rights Reserved, Grist.org
Otto, Nacin at Grist
Last week we were lucky enough to play host to two members of the WordPress core team last week here in our offices in Seattle. If you ever get the chance to work with/learn from Otto (@otto42) and Nacin (@nacin) … well just do it! Grist would not be what it is without the contributions of many people, but it many ways it’s contributions of time and energy by experts like Otto and Nacin who make doing what we do possible. So thanks guys.
More soon on what exactly these guys were doing with us last week, all we can say is that we’re excited!
Downtime on Grist.org Today
Hi all, we’d like to apologize for some downtime you may have experienced earlier today on grist.org. While the reasons for the downtime are somewhat arcane, and have to do with an outage experienced by one of our providers, we think it still probably made for a pretty lame experience for many of you on the site. Sorry about that.
If we actually made stuff other than sharp, snarky journalism (like frozen organic veggie-dogs or low-wattage lightbulbs?) we’d probably start handing them out for free as a way to say … um … our bad. But given that we are a website and a community, the best we can do is tell you that we’re sorry, that the problem has been permanently fixed, and that we do not expect any further interruptions.
Thanks, as always for reading Girst.
Update: As of 11AM EST on Thursday, October 27th, service should be restored to the vast majority of the world. If you continue to not be able to see the site, this condition is temporary and will resolve in the coming hours.
Halloween Infographic
Before you set out and conquor the zombies, demons, and princesses, have a look at some numbers about Halloween you may find interesting.
All Rights Reserved, Grist.org
We’re hiring!
Are you a news-oriented front-end developer, or know someone who is? Grist is hiring a developer. Details here.
pdfSpy
The other day at the Hacks/Hackers hackathon at ONA11, I collaborated with Angelica Peralta Ramos of La Nacion (Buenos Aires) on an interesting problem. When journalists find themselves covering large government organizations or departments, and when those departments (as is frequently the case) like to release source material by posting huge index pages consisting of links to PDF files or other static assets (like this!), a problem of tracking arises. What if the government decides to simply change one of those 400 documents in some possibly important way? Are journalists supposed to look at each document themselves each day to check?
This seemed like an obvious candidate for automation, so Phillip Smith and I discussed some ideas of how to automate the checking. The result is a simple script called pdfSPY, which, if cron’d will repeatedly check an arbitrary index page like this for changes to its linked files. The first time it runs the script goes out and downloads a local archive (at path-to-archive) of the files. On subsequent iterations, it checks for any changes and if necessary downloads new versions of the files, creating a historical record of changes. (It can also send the journalist an alert email … coming soon.)
Code:
https://github.com/mattoperry/pdfSpy
Useage:
python pdfspy.py url-to-index-page path-to-archive
So there you go -- never be fooled by "government transparency" again -- watch them with your pdfSpy.
Note: This code is rough rough rough. Please do with it what you want, but clearly I make no assertions that it will even work for any particular purpose.
