Sign in or Join FriendFeed
FriendFeed is the easiest way to share online. Learn more »

Julien › Comments

Julien
Re: It’s Official: AIs are now re-writing history - http://www.usv.com/posts...
"This is quite scary." - Julien
Julien
"Great shot!" - Julien
Julien
Re: Search Vs Social - http://avc.com/2014...
"Be really careful, because "Direct" is more of a "we don't know where it's coming from" than people typing the URL in their browser :) In the day and age of HTTPS, the Referer, which is used to measure this is now obfuscated, resulting in an increased of "direct" visits..." - Julien
Julien
Re: Julien Genestoux: Why RSS Needs a Follow Button - http://blog.flowreader.com/julien-...
"Thanks for accepting my guest post! Also, check a few sites with the button: - http://www.christophh.net/ - http://usv.com/ (click the feed icon!) And many, many more!" - Julien
Julien
Re: Radio3 v0.56 has a deeper API - http://radio3.smallpict.com/2014...
"It's not a deal-stopper but would make things a bit easier easier :) Many HTTP client libraries will parse the body by default when they *see* it's JSON." - Julien
Julien
Re: Radio3 v0.56 has a deeper API - http://radio3.smallpict.com/2014...
"Dave, this is a great feature (more to come on what I built with it). Since you're sening JSON, you should make sure the Content-Type indicates that it's JSON. Right now you're sending this: Content-Type: application/x-www-form-urlencoded; charset=UTF-8 Fix it by sending Content-Type: application/json; charset=UTF-8" - Julien
Julien
Re: The Old Reader: behind the scenes - Relevance - http://blog.theoldreader.com/post...
"I really hope you address this problem in a new way. What I really like with my RSS readers is that they do NOT filter. Facebook only shows what *they* think I care about. I'm sure there is a lot of false positive (just looking at the feed!) and I'm afraid there are a lot of false negative (stuff they don't show that I care about). However, I agree that "2619" is not a good UI/UX. There *must* be way to surface interesting storied or help me go thru them in more than one dimension? Maybe "fold" related stories so that I can go from one to another unrelated one (usually the case with chrnonological sorting) or from one to a related one (folded)... etc. So I think the problem is not algorithmic, but clearly an interface problem." - Julien
Julien
Re: The Slow Hunch - Your phone is no longer a part of you. It’s a... - http://nickgrossman.is/post...
"On the other side so many people have been "saved" when they found out that a backup of lost data existed in iCloud. It's a really hard problem to solve from a UX standpoint: we expect more from the computer without really understanding the consequences of these expectations." - Julien
Julien
Re: Digg Blog, New Today: Digg Deeper Goes Deeper - http://blog.digg.com/post...
"Well done Digg!" - Julien
Julien
Re: Turns out we still need Plaxo (or something like it) - http://josephsmarr.com/2014...
"All of this is very true, but I'm still not sure it's a "solvable" problem or at least that there is a much better solution than what we have today. To me, identities have facets and evolve over time very much. A "contact" is more and more a facet of an identity. It used to be that people had the same snail mail address their whole life, but our parents showed that this wasn't true anymore. We also change our phone numbers, emails probably more often than what we initially expected (or at least more than telcos expected) and web identities (profiles on social networks) are even more ephemeral (at least in their usage). Of course, smart phones and social networks should have make all that much easier, but in practice, these are all walled gardens will *less* interrop... If I were to build a tool that helps me keep track of my contacts contact info, I'd probably build a search engine. Something in which I could type a name (or a list of names), which would crawl the web, and every..." - Julien
Julien
Re: Superfeedr : Server Sent Events - http://blog.superfeedr.com/server-...
"So, for anyone reading this is the future: the headers approach does not work well in Firefox 31. It's much easier and safer to use the query string :)" - Julien
Julien
"It's worse than that. So many times I've heard people say: "I truncated my feeds because I need the visitors *on* my site". And then, a couple months later: "Nobody visits my site thorugh RSS, so I decided to drop them altogether". It's a sel-fulfulling profecy: if you make anything crap, don't expect people to love it more! That being said, I believe the RSS ecosystem should also start thinking about how to become more useful for *publishers*. There are 2 things that I see which are wrong: - It's soooooo hard to subscribe to a site that I love and usually involves copy and paste which are impossible on mobile/tablets. Luckily there is https://subtome.com (which I created) but it's not widely adopted. - I wish there was a way to "poke" the sites I read i their feeds. It would probabaly encourage them to improve their feeds and show them that I'm still there, reading them, even though they can't see me." - Julien
Julien
Re: How to stimulate the open web - http://scripting.com/2014...
"I strongly agree with your statement: " If I create a tool that's good at posting content to Facebook and Twitter, it should also post to RSS feeds, which exist outside the context of any corporation." I also wish people who rely on RSS would build tools to show publishers that the ROI with RSS feeds is greater than the one they get with Twitter/Facebook. For example, I'm always puzzled when I see that the most engaged users on Gizmodo, per Chartbeat are the ones coming from RSS reader and not the ones coming from Readers. In other words, us, defenders of RSS should make sure that its performance is measurable so that publishers can decide to push it forward rather than consider it as a 2nd tier citizen." - Julien
Julien
Re: The Slow Hunch - Dropbox and Personal Data - http://nickgrossman.is/post...
"Agreed, this is indeed an approach that I like much better. I'm always scared when data and code are mingled in an 3rd party entity without any way to de-tangle them. However, I think Dropbox could(should) do a lot better when it comes to privacy. For example, they could start encoding things..." - Julien
Julien
Re: Superfeedr : Server Sent Events - http://blog.superfeedr.com/server-...
"Indeed, that's a typo :/ Fixed now! I should try the username:password trick, but there's still a risk of it being ignored by browsers when the query string will *always* work :)" - Julien
Julien
"Happy Birthday!" - Julien
Julien
Re: The Dentist Office Software Story - http://avc.com/2014...
""it monetizes with native advertising" Yikes. In what world is everything supposed to be free and monetized with advertising? In the end, who will pay those ads if nothing is bought to pay these ads?" - Julien
Julien
Re: Google has run away with the web search market and almost no one is chasing - Quartz - http://www.usv.com/posts...
"This is very sadly true. I'm actually still amazed Amazon has not introduced a decent search engine (outside of Facebook.com search) which would use a different signal (probably some kind of social signal). Similarly, I believe we're entering an era where the link based signals (page rank!) is not the only option, in the same way that the SERP (result pages) are not the only way to display results." - Julien
Julien
Re: How many Internet users know what RSS is? - http://scripting.com/2014...
"1BN people use Facebook everyday, how many of them know what TCP is?" - Julien
Julien
Re: Where's Circa's RSS feed? - http://scripting.com/2014...
"+1 Hopefully Matt will listen and add that!" - Julien
Julien
Re: The Smart Watch - http://avc.com/2014...
"I'm convinced we're missing the point about "wearables computing". I just finished Nexus and I'm convinced we'll see embedded computing before we actually see masses adopt watches, glasses or other types of wearables." - Julien
Julien
"I am still not sure you were wrong, or at least, I'm still not sure what was wrong with Amy St and I'd love to hear the lessons. VHX looks very cool!" - Julien
Julien
Re: Queues and databases - Antirez weblog - http://antirez.com/news...
"So, at Superfeedr we use queues in the form of rings. The idea is that whenever an item goes into th queue, it stays there forever thanks to rpoplpush, and we actually distribute the work accross different servers using rings or various sizes (different number of elements): fastests rings have the least elements and slowest rings have the most elements. Our biggest problem was indeed cleaning, because rpoplpush will always push an item back, even if we do not need that item anymore. We do that by having other processes which empty and fill up rings, without prossing items." - Julien
Julien
Re: You gotta read this! / Thoughts about reading and internet media use - http://www.borthwick.com/weblog...
"I'll comment here because I can't do it properly on Medium. These 2 waves are definetly interesting and I do also believe they're both needed and probably feed one another. The "speed/bait/share" loop keeps us engaged with a higher level of attention and awareness, while the "long form/read/save" loop keeps us in a learning and curious state. It's almost like fast-food vs. Michelin-star restaurant. The former allows you to consume quickly and feed your "primal" needs, while the second helps you enjoy the act of eating and sharing time with friends. Now, though, I believe our current "tools" are not picking their sides enough. Of course, Facebook and other social network want to claim their bringing high value content to their users, when in practice they probably only deliever a "fast food experience" and on the other end, reading tools like feed readers, save for later, or even magazin style applications are probably struggling to show the volume and velocity that would make them..." - Julien
Julien
Re: Serious question: What makes the FB emotion study... - The Gong Show - http://thegongshow.tumblr.com/post...
"And finally, "product development testing process" applies to Facebook... not the data that's posted by my friends, unless of course, they consider that their products are, indeed, my friends." - Julien
Julien
Re: Serious question: What makes the FB emotion study... - The Gong Show - http://thegongshow.tumblr.com/post...
"Also, all the comparisons you make are not apples an oranges. When Gmail updates their interfaces with A/B testing, they do not hide or show specific emails. They at best change the interface around *all* of them, and at worse claissify them, but you can *always* see all the messages." - Julien
Julien
Re: Serious question: What makes the FB emotion study... - The Gong Show - http://thegongshow.tumblr.com/post...
"I strongly disagree with your post and I am extremely scared of what Facebook's been doing (is doing). The most obvious difference I can see between the optimization makes with "my feed" and a company makes with "their products" is exactly that: "my" vs. "their", or more important, 1 to many and 1 to 1. When P&G designs a product, it's not only for me. It's for milions (most of the time) people, which means that we can compare the effect of their producs among these people and see what exactly is delievered to *everyone*. The recipe of a shampoo is the same for you and I if we use the same shampoo. If you have an allergic reaction, it's easy to test/reproduce and see if anything is wrong with the product and when it is. When Facebook "optimizes" my feed, they do not otpimize Facebook. They optimize my feed. There is a difference between them changing the general UI, the colors or even the words in the interface and changing what I can see from my friends, or worse, what they want to..." - Julien
Julien
Re: A Tour of Input Methods - The Gong Show - http://thegongshow.tumblr.com/post...
"Very interesting analysis, but I can't wait for a follow up on input methods on mobile. I agree that you can probably make a difference with a better input, because the desktop/laptop is already optimized in terms of inputs. The mobile devices (including tablets!) are still extreemly far being even remotely has efficient and I'm convinced that if touch gesture are a huge progress over the mouse, we still have not found the equivalent for text inouts and keyboards are still very painful to use on mobile. Maybe voice will fix this, but I doubt it...." - Julien
Julien
Re: The Law Of Unintended Consequences - http://avc.com/2014...
"I don't want to wear the Kid's clothese, but isn't the subprime crisis also the consequence of policies established 20+ years before which aimed to helping *anyone* become a homeowner? In the end, hasn't this put more people on the streets?" - Julien
Julien
Re: Is Regional Currency A Viable Economic Tool? - http://eigenjoy.com/2014...
"This is a fascinating topic and once can be surprised to see that up until BTC, there has been so little innovation on the currency front, despite centuries worth of technical progress... yet, I feel like the risks associated with disruption in that area are so high that eventually, most people and organizations will just be "too scared" to change things." - Julien
Other ways to read this feed:Feed readerFacebook