root

[root]

Just discovered the queue, this thing is amazing

[root]

first rule of software development is just deploy that shit baby

[root]

So, I found a cool app that adds typewriter sounds to your keyboard and nothing strange happened :)

[root]

🖥

[root]

The Apple Lisa? It runs Xenix.

The Apple Lisa was the direct predecessor of the Macintosh line. It was developed after a team of Apple engineers and Steve Jobs toured Xerox PARC, who were hard at work developing the first graphical user interfaces on their Alto machine. While the Alto and the concepts it brought to the table were revolutionary, it was never a mass-market computer - only 2,000 units were shipped to customers.

The Lisa was Apple’s first attempt to bring the GUI to the masses. Aimed at the professional/business market, it was marketed at $9,995 (in 1983! That’s about $24,000 in today’s money) and it came bundled with a 5 MB hard drive. Unfortunately, poor hardware design choices and business decisions by Apple (most notably Steve Jobs announcing an upcoming machine which wouldn’t be compatible, but would be much cheaper and more powerful) led to the system’s discontinuation in 1985. It went through two hardware revisions and price drops and managed to sell 100,000 units.

Xenix was ported to the Lisa by Microsoft and SCO. It allowed terminals to be attached to the Lisa by serial ports and there was a text-based word processor available for it. Xenix did not support the Lisa’s graphics or the mouse. The disks are available online and have been confirmed to work with Lisa emulators.

Additionally, a version of UniPlus System V Unix has been reported to have been ported to the Lisa, but not much information on it is available and I have been unable to find any screenshots of it. Some documentation has surfaced, however. Notice how the manual uses “Macintosh XL” instead of “Lisa” - after the release of the Mac, a hardware redesign of the Lisa was sold under the Macintosh XL branding, ostensibly placing it as a “high-end” machine. It didn’t sell very well and was discontinued shortly after release.

[root]

What’s encrypting your internet surfing? An algorithm created by a supercomputer? Well, if the site you’re visiting is encrypted by the cyber security firm Cloudflare, your activity may be protected by a wall of lava lamps.

Cloudflare covers websites for Uber, OKCupid, & FitBit, for instance. The wall of  lamps in the San Francisco headquarters generates a random code. Over 100  lamps, in a variety of colors, and their patterns deter hackers from accessing data.  

As the lava lamps bubble and swirl, a video camera on the ceiling monitors their unpredictable changes and connects the footage to a computer, which converts the randomness into a virtually unhackable code.

Codes created by machines have relatively predictable patterns, so it’s possible for hackers to guess their algorithms, posing a security risk. Lava lamps, add to the equation the sheer randomness of the physical world, making it nearly impossible for hackers to break through.

You might think that this would be kept secret, but it’s not. Simply go in and ask to see the lava lamp display. By allowing people to affect the video footage, human movement, static, and changes in lighting from the windows work together to make the random code even harder to predict.

So, by standing in front of the display, you add an additional variable to the code, making it even harder to hack. Isn’t that interesting? 

via atlasobscura.com

[root]

Updating Tumblr on the Web

We posted an update about this back in April, and now as of July 1st, it’s really, really here, for everyone: the old dashboard has been replaced with our brand new web experience on desktop. This has been a very long time in the making, and the primary reason behind it is to make the desktop web experience of Tumblr easier to maintain and build on top of.

We’re continuing to improve the experience of using Tumblr on the web with some new features, some of which were formerly a part of XKit and other third party extensions:

Color Palettes are now available to change the whole look of the site, just use the “Change Palette” option by clicking on the silhouette icon at the top right.

Viewing tags used in reblogs is now available in the notes view on every post.

You can now filter posts by their text content, not just by tags.

Timestamps are available by hovering over the “fold” at the top right of any post, or now also available by clicking on the meatballs menu at the top right of any post. There are a lot of new options in there, too!

The dashboard now soft refreshes by default, so you don’t have to press that browser button to see the latest content.

Audio in audio posts can now “pop out” so you can see it while you scroll your dashboard.

There’s now a CSS map, API access helper, and more, available to third party extension developers. Keep an eye on this repository!

One piece of feedback we heard a lot was allowing pagination by changing the URL of the dashboard, and that’s something we plan to support. Thank you again for all of the insightful feedback about the new web experience, keep it coming!

[root]

Dusting off the cobwebs

Over the past couple of weeks, we have slowly rolled out a spiffy new version of Tumblr on the web. Here’s a smattering of goodies in the update:

Dark mode is now available on the web! Along with a few other color schemes, too. Just click the silhouette figure in the upper right corner and then click “Change Palette” to cycle through all of ‘em.

You can now view your likes page in the default view or in the new grid view.

Same with any tagged pages. Here, try it out with the frog tag. A glorious grid.

Copying the link to a post, reporting a post, or blocking someone whose content you don’t enjoy is easier than ever. Just click the meatballs menu (●●●)  in the upper right corner.

Some of these visual changes are obvious, but the biggest transformation is the one you can’t see. This wasn’t just a facelift, we completely updated the web interface. The old one was outdated—over a decade old. Adding features to it wasn’t always easy, fixing bugs wasn’t always quick. It was clear that we needed to update the whole dang thing. So update the whole dang thing we did. The result? A faster Tumblr. The experience is smoother, it’s easier to add new features, and we’re able to squash any pesky bugs quicker than before.

XKit

This kind of web interface change means that some of your Tumblr browser extensions may not work. We understand how important these extensions are to those of you who want to customize your Tumblr experience as much as possible. Throughout this process, we’ve stayed in close communication with the people over at @new-xkit-extension to make sure our update was one that they could work with. We provided them with APIs they needed to begin moving over to this new platform. This also gave us the opportunity to discuss the benefits their extension brings to the community. Having their input on what makes their service valuable to the people who use it has been a true treat, and one we are grateful for. While XKit isn’t an official Tumblr tool, it’s one we know is dear to many of you, and we don’t take that lightly. If you have any concerns, we encourage you to read more about what they have to say over here. Be sure to follow them to stay updated on their process!

We want to hear from you

Yeah, you. When we first began testing this new interface with a small selection of you, we received some really useful feedback. We want to make sure that continues, so we created a survey. While we always read the reblogs on our announcement posts, this survey is the best way to let us know exactly how you feel about this update. And don’t worry—it’s anonymous. We won’t know who you are unless you make it your business to tell us so. Head on over there and share your thoughts now.

[root]

Windows 7 is due to reach EOL today. However, a large number of the systems, most in corporate environments, are still running the Win7. So I decided to advance my agenda of pushing the Linux desktop during the conversation. It went to /dev/null. O well, I tried my best.

[root]

Advances in Spam Detection on Tumblr

As with all open platforms for user-generated content, Tumblr has been hit with a fair bit of spam. People who create spambots or abuse our platform in the interest of non-genuine social gestures are really good at finding new ways to develop and implement their spam. It’s what they do. Over the years, we have been experimenting with various tools and techniques to combat issues like spambots and non-genuine social gestures. To understand more about our work, let’s dig into the details.

Challenge: precise identification of spam

Spammers often try to disguise themselves by attempting to use a platform in the same way a real person would. As spammers learn how to develop newer, better ways of mimicking the behavior(s) of real people, the boundary between spammers and real people becomes more and more blurred, which unfortunately means non-spammers may get flagged as spam. This is what is known as a false positive. 

Tumblr’s goal has always been to find the delicate balance needed in making sure we are addressing spam as aggressively as possible without dramatically increasing the number of false positives.

Challenge: evolving spam behavior

Spam evolves. Spammers learn how to dodge new spam detection as soon as a platform starts using it. Therefore, relying on fixed logic is not sufficient. We instead approach the issue with a broad set of dynamic predictors, because the best way to combat spam is to utilize an adaptable detection methodology.

Our work

At the heart of all good spam detection efforts are machine learning algorithms. These algorithms are fed data from how real people use Tumblr and use this data to enhance our classification accuracy. Thanks to this historical data, when new spam or malicious patterns start occurring, we can react faster and identify spam with higher accuracy. Our newly launched model demonstrates 98% accuracy in determining if a user is a spammer.

The diagram below describes our spam classification pipeline:

Because every machine learning algorithm starts with data, we begin with a data management system that manages and controls data streams flowing around Tumblr. Every microsecond, the data management system records this data into log files. The system then periodically transfers these logs into our database, or Hadoop File System (HDFS). We then write numerous scalding jobs that focus on identifying what parts of this mountain of data are helpful when learning who’s a spammer. To start this process, we come up with specific hypotheses on some data sources and then collect the data to test these hypotheses in the next step.

How do we test if a data source is useful? After the scalding jobs finish, we analyze and visualize the data source to determine if the collected information can be turned into a signal. If the raw data itself is not enough, we might need to combine several signals to produce better results. The whole thing may sound a little hard to grasp for some. Maybe a pseudo example could help?

What if we found that many spammers really enjoy, say, insects and they were creating posts with a massive number of insect images? Based on this observation, we would hypothesize that the more pictures of insects someone’s blog has, the more likely it is spam. If we validated this hypothesis, we would then build a feature called InsectImageNums to track how many times a blog has posted an insect. But wait! What if we realized that the majority of our users post zero insect images? This becomes problematic because most of the data in InsectImageNums are zeroes, and those that are not zero have a very diverse range. Besides, some insects specialists or nature lovers do post images of insects, and we don’t want to classify these people as spam incorrectly. We would need to dig deeper and find a more detailed differentiator. Perhaps we see that it is rare for even the most bug-loving person to post more than five pictures of insects. We’d use that finding and create a new predictor called InsectsImageNumsGreaterThanFive. After this transformed feature is verified as accurate and useful, it is included in our predictor set.

When we have a verified and helpful set of working features, we then pass them to the machine learning models in Spark through Hive. Sometimes the aggregated size of the data is way too big for a single machine to process, so we use Spark and Spark ML interface to train our larger data-sets. 

What kinds of machine learning algorithms are we using? 

Supervised machine learning requires training labels, but these labels are only partially defined. With imperfect labels, we use iterative semi-supervised machine learning techniques to label instances closest to the classification decision boundary by checking our predictions with human agents. When human agents stop seeing false positives, we assume the model is crafted strong enough to be placed into HDFS. Through this semi-supervised approach, we achieve a 98% accuracy rate. We then upload the trained machine learning model to our database and periodically update it.

We save the spam probability score of new groups of users daily on Redis, an in-memory data structure store. This user spam probability score becomes a useful data validation point for our internal team that leads our spam moderation effort. In a way, the machine learning spam detection pipeline’s job is not to automatically suspend suspicious blogs, but to find blogs that have suspicious behaviors—like spreading viruses or malicious content across the internet. We want our community to enjoy a friendly environment on Tumblr, and we want to avoid as many false positives as possible. That’s why our overall pipeline involves both machine and human efforts.

What’s next 

Spam detection work is never done. What works above may not work six months from now. Our goal is to evolve one step ahead of the spammers. Keep your eyes peeled here on @engineering to stay up-to-date!

— Vincent Guo (@dat-coder)

[root]

margin-left: -16px;

[root]

[root]

JS logic

[root]

One of my better comebacks.

[root]

Perks of open source.

[root]

It’s that time of year again

[root]

my prof asked us to evaluate the graph of the laplace of 1/t... like is this a trick question the integral of 1/t * e^-st does not converge for t=0..inf