Firefox OS'lu Mozilla Flame satışa çıktı
Kimden mi bahsediyoruz: Tabii ki Mozilla! Aradan geçen zaman içerisinde Firefox OS'u geliştiren Mozilla, uzun zamandır beklenen ve bu işletim sistemi ile çalışan telefonunu nihayet piyasaya sürdü! Firma tarafından yapılan açıklamaya göre Flame isimli ...
Mozilla'nın İlk Telefonu Flame Satışta!ShiftDelete
Mozilla ilk telefonunu satışa sunduTimeturk
alle 17 nieuwsartikelen »
If you discover my project: this is a browser which is controlled by a script, not by a human. So it has no user interface. In fact this is a browser like PhantomJS, which proposes the same API as PhantomJS. But it is based on Gecko, not on Webkit. See my previous post about the start of the project.
This new version fixes some bugs and is now compatible with Gecko/Firefox/Xulrunner 31.
Next big work on SlimerJS:
- fix last issues that prevent GhostDriver to work well with SlimerJS
- support Marionette(https://developer.mozilla.org/en-US/docs/Mozilla/QA/Marionette)
- try to implement remote debugging, to allow to debug your script from Firefox Dev Tools
- try to have a true headless browser (so to have a browser without visible windows)
Help is welcomed, See you on Github ;-)
About 2½ years ago I wrote a demo for Mozilla Hacks how to use Canvas to create thumbnails. Now I felt the itch to update this a bit and add more useful functionality. The result is:
It is very easy to use: Drop images onto the square and the browser creates thumbnails for them and sends them to you as a zip.
You can set the size of the thumbnails, if you want them centered on a coloured background of your choice or cropped to their real size and you can set the quality. All of this has a live preview.
If you resize the browser to a very small size (or click the pin icon on the site and open a popup) you can use it as neat extra functionality for Finder:
All of your settings are stored locally, which means everything will be ready for you when you return.
As there is no server involved, you can also download the app and use it offline.
The source, of course, of course is available on GitHub.
To see it in action, you can also watch the a quick walkthrough of Makethumbnails on YouTube
August 18, 2014 we will carry out the webmaker event we’ve scheduled previously, the event held at SMK ITACO Bekasi, this is a vocational school for children who are less economic conditions. We only...
There’s been a couple of developments with the social networks Facebook and Twitter that fit together quite nicely this week. The first is the news that Facebook likes make a huge difference in terms of what you see while browsing your news feed:
Wired writer Mat Honan found out what happens when you like every single thing that shows up in your Facebook feed. The results were dramatic: Instead of his friends’ updates, he saw more and more updates from brands and publishers. And, based on what he had liked most recently, Facebook’s algorithm made striking judgements about his political leanings, giving him huge numbers extremely right-wing or extremely left-wing posts. What’s more, all that liking made Honan’s own posts show up far more in his friends’ feeds — distorting their view of the world, too.
But Medium writer Elan Morgan tried the opposite experiment: Not liking anything on Facebook. Instead of pressing like, she wrote a few thoughtful words whenever she felt the need to express appreciation: “What a gorgeous shock of hair” or “Remember how we hid from your grandmother in the gazebo and smoked cigarettes?” The result, as you might guess, is just the opposite of Honan’s experience: Brand messages dwindled away and Facebook became a more relaxed, conversational place for Morgan.
The second piece of news is that Twitter is experimenting with changes to the way that ‘Favorites’ work:
Favorites have also been pseudo-private; while you can view a list of favorited tweets from an account’s profile page or on a tweet’s detail page, typically only the “favoriter” and the “favoritee” ever know about it. If Twitter starts surfacing favorited tweets in timelines, they’ve suddenly become far more public. The change — and the backlash — is somewhat similar to Facebook’s attempts to share just about everything “friends” did with Open Graph.
For those who have used Twitter for years, the change is so shocking it can seem like the company is completely ignorant to how its customers use the service. But even seasoned Twitter veterans should admit that the service’s core functionality is fairly arcane — it’s far from accessible to new users, and that’s a problem for Twitter.
What I find interesting is that most sites allow you to ‘love’, ‘like’, ‘favourite’, ‘+1’ or otherwise show your appreciation towards content. You can do this with Mozilla Webmaker too, when browsing the gallery. The trouble is that this is extremely limiting when it comes to data mining. If it’s used in conjunction with an algorithm to serve up content (not currently the case with Webmaker) then it’s a fairly blunt instrument.
There are some sites that have attempted to go beyond this. I’m thinking specifically of Bit.ly for Feelings, which allows you to share content that you don’t agree with. But there’s not a lot of great examples.
The trouble is, I guess, is that human emotions are complex, changeable and along three-dimensional analogue spectrum. Digital technologies, on the other hand - and particularly like/favorite buttons - are binary.
Update: after posting this I found that Yahoo! are planning to scan photos you publish on Tumblr to gauge brand sentiment. I’m not sure if that’s better or worse, to be honest!
My colleague asked me to take a look at a logging issue on a server last week. He noticed that the error logs had way too little information about exceptions. In this particular instance, we had switched to Nginx + gunicorn instead of our usual Nginx + Apache + mod_wsgi (yeah, we’re weird). I took a quick look this morning and everything looked exactly like they should. I’ve read up more gunicorn docs today than I’ve ever done, I think.
Eventually, I asked my colleague Tryggvi for help. I needed a third person to tell me if I was making an obvious mistake. He asked me if I tried running gunicorn without supervisor, which I hadn’t. I tried that locally first, and it worked! I was all set to blame supervisor for my woes and tried it on production. Nope. No luck. As any good sysadmin would do, I checked if the versions matched and they did. CKAN itself has it’s dependencies frozen, this lead to more confusion in my brain. It didn’t make sense.
I started looking at the Exception in more detail, there was a note about email not working and the actual traceback. Well, since I didn’t actually have a mail server on my local machine, I commented those configs out, and now I just had the right Traceback. A few minutes later, it dawned on me. It’s a Pylons “feature”. The full traceback is printed to stdout if and only if there’s no email handling. Our default configs have an email configured and our servers have postfix installed on them and all the errors go to an email alias that’s way too noisy to be useful (Sentry. Soon). I went and commented out the relevant bits of configuration and voilà, it works!
Image source: Unknown, but provided by Tryggvi :)
I am excited to announce that WebIDE is now enabled by default in Nightly (Firefox 34)! Everyone on the App Tools team has been working hard to polish this new tool that we originally announced back in June.Features
While the previous App Manager tool was great, that tool's UX held us back when trying support more complex workflows. With the redesign into WebIDE, we've already been able to add:
- Project Editing
- Great for getting started without worrying about an external editor
- Project Templates
- Easy to focus on content from the start by using a template
- Improved DevTools Toolbox integration
- Many UX issues arose from the non-standard way that App Manager used the DevTools
- Live memory graphs help diagnose performance issues
All projects you may have created previously in the App Manager are also available in WebIDE.
While the App Manager is now hidden, it's accessible for now at about:app-manager. We do intend to remove it entirely in the future, so it's best to start using WebIDE today. If you find any issues, please file bugs!What's Next
Looking ahead, we have many more exciting things planned for WebIDE, such as:
- Command line integration
- Improved support for app frameworks like Cordova
- Validation that matches the Firefox Marketplace
Mozilla has a number of source repositories under https://hg.mozilla.org/hgcustom/ that cumulatively define how version control works at Mozilla.
The latest addition to this repository is the import of the hghooks repository. This now-defunct repository contained all the server-side Mercurial hooks that Mozilla has deployed on hg.mozilla.org.
Soon after that repository was imported into version-control-tools, we started executing the hooks tests as part of the existing test suite in version-control-tools. This means we get continuous integration, code coverage, and the ability to run tests against multiple versions of Mercurial (2.5.4 through 3.1) in one go.
This is new for Mozilla and is a big deal. For the first time, we have a somewhat robust testing environment for Mercurial that is testing things we run in production.
But we still have a long way to go. The ultimate goal is to get everything rolled into the version-control-tools repository and to write tests for everything people rely on. We also want the test environment to look as much like our production environment as possible. Once that's in place, most of the fear and uncertainty around upgrading or changing the server goes away. This will allow Mozilla to move faster and issues like our recent server problems can be diagnosed more quickly (Mercurial has added better logging in newer versions).
If you want to contribute to this effort, please write tests for behavior you rely on. We're now relying on Mercurial's test harness and test types rather than low-level unit tests. This means our tests are now running a Mercurial server and running actual Mercurial commands. The tests thus explicitly verify that client-seen behavior is exactly as you intend. For an example, see the WebIDL hook test.
So what are you waiting for? Find some gaps in code coverage and write some tests today!
What are we learning? This post highlights new metrics and some early analysis from Adam, Amira, Geoff, Hannah and many others. The goal: turn our various sources of raw data into some high-level narrative headlines we can learn from.
Current contributor count: 5,529 (Aug 15)
- Are we on track to hit 10K? No, not yet. The statistical increase we’re seeing is based on good work to record past contribution. But our current growth-rate isn’t enough.
- Why is the 4-week trend-line up? Because of Maker Party + bulk capturing historical activity (especially Hive + MVP contribution badges).
- What can we do to grow faster? Short term, we can focus on (amongst other things):
- 1) Maker Party partners. Convert more partner commitments into action, through a streamlined on-boarding process.
- 2) Webmaker users. Try to convert more users into contributors. Ask them to do something more directly.
- 3) Training. Net Neutrality teach-ins, train the trainer events, MozCamps, etc.
- + …what else?
- We now have about 120K Webmaker users. We’re seeing big recent increases, mostly thanks to the snippet.
- About 2% of those users are currently contributors.
- ~50% of users have published something.
- Most of that publishing happens on the user’s first day. (Users who don’t make something on their first day tend not to make anything at all.)
- There’s very little overlap between tools. Users tend to make with a single tool. (e.g., of the ~46K people who have made something, only 2K have made something with both Thimble and Popcorn.)
- About 20% have opted in to receive email updates from us. (e.g., via BSD)
- Our top snippet performer: “The Web is your playground! See what you can build with Mozilla Webmaker and our global Maker Party.” (+ animated pug icon)
- CTR = 0.58%. (Other MP variations: 0.15% – 0.49%)
- The icon and animation have a big influence on CTR. Fun icons and playfulness are the hook.
- “Teach and learn” language generally performs as well as more playful language.
- Our top snippet performer: “The Web is your playground! See what you can build with Mozilla Webmaker and our global Maker Party.” (+ animated pug icon)
- Landing pages
- A “survey-based approach” is our top performer. Asking people *why* they’re interested in Webmaker. (vs straight email sign-up ask) (+4.7% conversion rate)
- 80 / 20 split for learning vs. teaching. About 78% of survey respondents express interest in making / learning, with 22% wanting to teach / mentor.
- Language focused on teaching, learning and education performs well.
- e.g., “Welcome to Webmaker, Mozilla’s open source education project, where you can teach and learn the web through making.” (+17%)
- vs. “We believe anyone can be a tinkerer, creator, builder of the Web. Including you.”
- Mozilla.org referral traffic
- “Webmaker” out-performs “Maker Party.” Our conversion rate dropped to half when we shifted from from “Learn the web” to “Join our Maker Party.”
“The further away we get from the Mozilla brand, the more work there is to get someone on board.” — Adam
- 1,796 events currently entered (Aug 15)
- That means we’ve already surpassed last year’s total! 1,694 total Maker Party events last year, vs. same number in our first month this year.
- But: we’ll still need a big event push in second half to hit our contributor target.
- Key takeaways:
- Tracking partner activity. Automated tracking has been hard — we’re relying instead on one-to-one calls.
- We’re gathering great data from those calls. e.g.,
- Unreported success. Partners are participating in ways that aren’t showing up in our system. Manual badging is filling that gap.
- Occasional confusion about the ask. Some think “Maker Party” is a “MozFest-level” commitment. They don’t realize the ask is simpler than that.
- They need easier ways to get started. More simplification and hand-holding. Working on a simplified “Event Wizard” experience now.
- Some partners see more value in Maker Party than others. Orgs with offerings similar to our own may perceive less value than those in adjacent spaces.
- We haven’t cracked the earned media nut. Not much coverage. And little evidence of impact from the coverage we got.
- We don’t have a good way for measuring participation from active Mozillians.
- Second half. We should gear up for a second “back to school” wave to maximize contributors.
“There’s the ‘summer wave’ and ‘back to school’ waves. We need to have strategies and actions towards both.” –HannahNext steps
- 1) Partner conversion. This is probably our best immediate strategy for boosting contribution. Ship a simplified on-ramp for Maker Party partners. A new “Event Wizard,” simple start-up events, and user success support.
- 2) Convert Webmaker users to contributors. We’ve seen a *big* increase in user numbers. This opens an opportunity to focus on converting those users. Ask them to do something more directly. Try new low-bar CTAs, email optimization, re-activating dormant users, etc.
- 3) Training. Train the trainer events, MozCamps, MozFest, etc.
- Year-long engagement. How do we more evenly distribute event creation throughout the entire year?
- Match-making. How do we identify the teachers? How do we connect those who want to learn with those who want to teach? What are the pathways for teachers / learners?
- Impact. How many people are learning? How much are they learning? Should we make “number of people learning” Webmaker’s KPI in 2015?