> According to Borthwick, it would have cost “hundreds of thousands per month” to keep the site running on its old platform. Even though the site was state-of-the-art just a few years ago, most of the infrastructure would be considered legacy technology by a modern startup.
why does it even need to keep on running? Geez, just keep open the submission process, crank up the ads, and turn off everything! Digg has no clue what passive income means. They should take a page from spammers.
I recall when Kevin Rose was interviewed prior to v4 launching, they were running the entire thing on Cassandra.
When it launched, they spent the majority of the time attempting to scale it in a quick fashion - a process which probably involved buying lots of resources and attempting to scale as fast as possible. It was badly designed from the start, and was never really fixed.
I was under the impression Facebook replaced Cassandra in their stack with HBase.
Yes, I forgot about Netflix! Not only do they use it, but they've provided tons of excellent tools for it (Astyanax, Priam) and blog about it very frequently.
It's not necessarily that, but when v4 launched, they experienced a giant problem with their infrastructure. Likely the "fix" was to increase performance versus fix the problem programatically.
Hence their previous statements. Kevin's statement was in reference to them switching completely to NoSQL and not knowing how to scale it properly.
To be fair, most of the stories on the front page of digg before this most recent redesign had 0 comments anyways, so they probably realized they could push out that feature a bit.
1) By using FB, at least initially, they can lower signup friction and get a good bit of social momentum. Especially if users are sharing their "diggs" publicly. Doesn't appeal to hackers, but it works when done tastefully.
2) Discussion on digg was always a bit janky. HN really spoils the idea of social links/comments because they are so high quality here, but when you're targeting the average person, its harder to control quality. Let people sound of on Twitter/FB, and you've created a new marketing vector without directly polluting the experience on your site.
While I haven't looked into this particular scenario. You used to be able to revoke individual permissions per app / site on FB. It's under Privacy > App & Sites, obviously it defaults to on as it's convenient for the site in question. An entirely different debate.
I think for a lot of people, being able to keep a pseudonymous identity is one of the few things that differentiates sites like reddit and digg from Facebook and twitter in the first place. Most people I know aren't crazy about letting others know what their reddit username is, let alone slapping everything they do on their Facebook feed.
Exactly. My id here is my first name and initial. Correlating my username here with my full name, which is unique worldwide, is easy.
My Reddit id on the other hand does not directly tie to my name, and that's just how I like it.
It's still fairly easy to find - I don't care if people who really want to figure it out, as I don't write anything I can't stand for. If I wanted true anonymity I'd take a lot of extra precautions.
I just care that my Reddit comments don't show up on the first few pages of a Google search for my name, as it lets me not think as much about my "professional image".
Putting it on Facebook is something I'd never let happen.
There's a huge difference between being possible to find what I've written if you really care about it vs. it being showed in the face of my family, for example.
And this is without being part of any controversial sub-reddits. There are plenty of sub-reddits or individual threads that are offensive enough or controversial enough that I'd imagine they'd die pretty quickly if people were forced connect their Reddit identities to anything that might be directly matched to their names.
From the FAQ Under 'Why Facebook [for logins]?' "We're building the Digg for 2012" - I get that like 1/13 people on the planet are on Facebook, but what about the other 12? Twitter? Or even shudder Google?
My lawd man, build your own. You restricted yourself to 6 weeks, not users.
Some discussion in the blog comments about that [1]:
>Using Facebook for account registration is a short-term solution that will seriously cut down on spam, while we take our time to develop more robust spam-filter technology. We know this isn’t ideal, so rest assured: we are working towards a more lasting solution. [2]
I guess Digg is aimed more towards general crowd now that it only allows for login with Facebook. In thier FAQ they mention it is to cut down on spam but something tells me we will never see User registration. What about people like me who do not have Facebook. I certainly do not want to make a Facebook account to use this service. Furthermore, there are a _lot_ of people who do not use it, why did they not consider this?
Maybe they'll just stick to the hundreds of millions of Facebook users, at least for now. The intersection of the sets of people that don't want a Facebook account and people that want a Digg account is probably small.
Furthermore, there are a _lot_ of people who do not use it, why did they not consider this?
"A lot" is a very relative term. Non-Facebook users are probably in a minority in the kind of audience they're targeting. I think it's an interesting choice. I'm not going to say it's sensible just yet, but when they introduce comments it will be interesting to see if forcing real names brings the level of discussion up.
"Using Facebook for account registration is a short-term solution that will seriously cut down on spam, while we take our time to develop more robust spam-filter technology. We know this isn’t ideal, so rest assured: we are working towards a more lasting solution."
So they lost the ability to comment and moved further away from the reddit/HN tight text view (which they used to look like and have been steadily moving away from) and now look like a messy news page.
Is anyone else finding this new redesign even more irrelevant than ever? What does digg provide at all now? No discussion and a very messy hard way to even find anything.
Let's say I were to redesign Ebay as a big picture listing site (ability to bid and buy, coming later)
Would anyone consider this a ready to launch replacement? If they don't value commenting as one the of the most core features what hope do they really have?
To me HN is more valuable for the comments then the aggregation. I often skip the article and just look at the comments. Digg cannot now compete with this and if this is their priorities: big gaudy pictures, voting, and distant third or less, commenting... well.
good luck I guess but we have different priorities on what we were hoping for and expecting and need.
I think that "launch" is a relative term here. I mean that in the sense that Betaworks have been entirely open about the process from the start, and are aware that 90% of their audience right now is curious techies like ourselves. I very much doubt that they'll be promoting it much.
This is an alpha, or a beta, or whatever you want to call it.
The whole "mason" best fit I suppose. In comparison to say reddit and HN, no real linear flow of headlines. Your eyes tend to dart all over the place with this design looking for information. The F pattern seems to be proven as the best layout for information absorption http://www.useit.com/alertbox/reading_pattern.html
Those inline tweets are grating. The overuse of lines to delineate related content looks visually noisy. I also think the top story is given way too much space. On my 1280 x 800 display, the new Digg shows me 2.5 stories. I'd like for them to use a tighter grid so they could fit 4-5 stories above the fold on lower resolution displays.
Yes! What's not to like about it, I find the content engaging, the photos (a la Zite) interesting and the site layout looks and works well. I don't see why all these naysayers are complaining about the masonry layout. Magazines looked that way for several reasons, one of them is people seem to like the format. I'm all for sites rethinking the classic list design as we sees here in hn, it's informationaly dense (hacker news) but that only works when the domain is very specific. For Digg, the magazine look works much better. Well done Digg for having some design flair.
I love the new design, and I love the new content. There were at least three articles there that caught my attention right away, that I stopped to read before even returning to this thread.
Content-wise, it's not the same ol' garbage that keeps getting regurgitated on Reddit, or the same ol' subjects that keep getting rehashed to death on "Hacker"News. ("PHP is a terrible language!" "No it's not!" "Yes it is!" ... yuk.)
Visually, it's not a list of links which can only draw your attention by gaming the headlines. It's about time someone re-remembered that a picture is, indeed, worth a thousand words.
As far as comments go, it's been a long time since I've been consistently enriched by reading the comments on anything anywhere. (This thread is a good example, my blathering included.) I've got my fingers crossed for Web 3.5: "Here's a nice page, shut up and enjoy your stay."
I had the same reaction: yay fresh content! Although, it's a quite a but early to say whether or not the content mix will become stale.
My biggest complaint is the layout that seems to assume a huge screen or something wider than my laptop has. Scrolling left-right in addition to top-down is annoying enough for me to bounce.
Hmm. On my setup, at least, the content container is 1045px wide ... not too wide, I would think. However! It looks like if you're zoomed in, you'll get the same font-size set to a wider page width, which isn't optimal. Is it possible you're zoomed in a bit?
If you honestly think HN is a model of good design, you're a purist in the worst sense. HN isn't even responsive. I love this place, but that's absolutely ridiculous. They might as well throw in a free AOL disk.
I think they needed to take away the ability to comment.
Digg was a pretty awesome community in 2005 and 2006. The content was largely tech related and the comments were informative and interesting.
In 2007 it started getting worse, there were more jokes and memes and the content started to suffer. With no subreddit like system the original community began to become drowned out.
Towards the end of 2008 there was nothing left except memes and lame one line jokes that put youtube comments to shame. I was always surprised by all the people complaining about Digg v4 - the community sucked a long time before that.
I'm impressed by this revision, should be interesting to see what happens.
To add to that, one more issue that caused me to drift away from Digg was the numerous posts/comments on sensitive topics such as religion, politics etc.
Reading the comments on the blog, I can completely understand why commenting wasn't the first thing to be implimented. They probably wouldn't put it this way, but I think they needed time for some of the less desirable user base to flee the site in order to start up again and be respected. I remember what happened after v4 was launched, and you couldn't read the comment section of any story without it being about the redesign. If you have a bunch of people complaining about a redesign instead of focusing on the content first, no one's going to even consider coming back, let alone joining for the first time.
It's no secret that the community needs a reboot, and content-first development might work for a site that through any other implementation might just strike people as a poor version of Reddit (regardless of whether or not that may be true).
This looks very promising. Seems like it'd be a great place to catch up on top stories, sort of like Flipboard. Curious to see what they end up doing with commenting.
The team behind this is clearly very skilled. Will be keeping an eye on this as it evolves.
So they really did nuke the whole thing. There's no legacy content migration, none of the user DB is intact. It seriously was just a premium domain buy. Fascinating
I'm all in favor of rapid releases and MVP, but this seems like a step in the wrong direction. As many have pointed out, the site is missing many of the core features required to build a community. This seems like a great way to alienate the few remaining Digg loyalists
I'm very interested to see what happens with this. I think they've made all the right moves so far since the acquisition and I love the transparency. They're going to try to stay true to their roots while also becoming "fast and thin", in their words.
It also makes a ton of sense to have the algorithm factor in tweets, FB shares, and diggs. Nice. I don't know if it'll ultimately work, but it looks like a good start.
Goddammit, I hate it when sites intercept middle-clicks. Middle click means "open in new tab", not "run some badly-written open-in-new-window javascript".
There's something about them calling it "Digg v1" that really annoys me. Digg version 1 was... the original Digg, no? I dunno what particular version of Digg this thing is, except that as a reboot of an existing property/brand it's obviously not version 1.
It might be their version 1, but calling it "v1" in public makes it feel like the old Soviet pictures where the leaders who had fallen out of favor were airbrushed out (http://en.wikipedia.org/wiki/Censorship_of_images_in_the_Sov...). "Nyet, Comrade! Commissar Rose was never in charge of this tractor plant!"
It just feels really disrespectful to the original Digg team, at least to me.
And it then went on to several more version over the years, so even if it started at 0 (which is ridiculous, this is version numbers not arrays) it would still end up at more than v1.
I think they mean v1 as in not beta or mvp. Also I have to admit it's good! However it is a waiting game now to see what sort of community builds around it.
It's a reboot of the franchise. Just like the new Star Trek movie wasn't called "Star Trek 11" but simply "Star Trek." It would be odd to keep the same version numbering scheme when they threw away the old system, no?
Oh no, not that "Magazine Layout" crap. I get why designers choose it, they want to make the page look lively and balance content all over the place but it's terrible for readability. It breaks natural eye flow from left to right then top to bottom. With magazine style layouts, your eyes have to go all over the place to skim through the content and that just gets annoying. If they wanted to avoid looking like a blog but still keep the lively look and feel they could have tried a grid like layout where at least the stories are aligned horizontally in a straight line.
The style itself is really nice, blocky and beautifully simple. This is how you redesign a content site, its elements are purposely made minimal, simple, and sterile so that the content stands out above everything else. The site disappears and the content takes the stage. Basically, it puts the spotlight on the performer rather than the stage. Also, the site's so lightweight that it just feels fast. Very few elements are begging for your eye's attention which is great, it allows the eyes to go straight for the image and title. Everything about the redesign is just so right, it's a shame they went for the magazine style layout.
Either way, great redesign. It's won me over, and I will actually start returning to digg to see how things go over the next few months. This was a very nice rebirth, back to simple, back to content.
The fundamental issue with the "magazine layout" is that it is completely pointless in a digital world. The reason why magazines are laid out the way they are is because there's a limit on space, so everything has to be crammed in wherever it'll fit. When you've got a website though, there are no such restrictions, so the focus should be on presenting content in the most orderly manner.
well to be fair there is some constraint in how much of a site afresh impression is going to look at and the average screen size they will be using. The grid is a less applicable but there's still some advantages to displaying all of your important bits in a semi clean fashion on the landing page with out flowing to far in any direction.
I personally like magazine style layouts, but the layouts need to be context sensitive like you would see in a magazine or newspaper. Stories that are more prominent in the design should be more popular. Stories should also be grouped sensibly.
Zite does this extremely well. Pulp, on the other hand, does it pretty poorly. Doing this on Digg's homepage is going to be challenging to not look schizophrenic. But on Digg's sub-topics, this type of layout will work well. I find scanning through tabular data to be fatiguing and I often scan over items out of boredom.
A 2d (versus 1d) layout is great for items that are meant to be browsed without order. It isn't an RSS feed or your inbox where you're supposed to diligently consume each piece before moving onto the next one in chronological order.
On my iPad, magazine layout via Flipboard is the only way I consume my content now. Because it is non-linear like Google Reader, I actually pay attention to each headline rather than skimming for key terms or phrases.
This layout for digg fits in with casual consumption and discovery, rather than a structure that encourages seeking out specific articles.
I do agree with you that this is a beautifully simple site and will also revisit on a regular basis to see how it progresses.
The right column is the one that is confusing, looks like its a place for top news or something, however it's just another column with the same relevance, this is a little weird, just content accumulated with no context. Also, why is there no bury button on the home page, having only [digg] = [like] is boring. Overall this looks like a wordpress theme for a magazine, very tired design. I would expect more from a digg redesign.
I have a facebook account but I have "deactivated" it (not deleted). I cannot use the new digg. The facebook popup shows up and then immediately disappears. This is frustrating.
>According to Borthwick, it would have cost “hundreds of thousands per month” to keep the site running on its old platform. Even though the site was state-of-the-art just a few years ago, most of the infrastructure would be considered legacy technology by a modern startup. Because of this, the new Digg team decided to throw away virtually all of the old underpinning of the site in favor of a fresh start. Borthwick wants to rebuild the company and to do so, he says, it’s important to turn it back into startup mode and develop a completely new and modern platform to develop the new Digg on.
Anyone know the tech stack they're using? Don't see anything specific on their website (blog, about, careers).
It began with PHP - I'm not sure if it was ever rewritten. However, by "the end" they were scaling with cassandra, but mostly they were throwing more hardware and people at it. Note that this is from a hazy memory.
The cost of running the site seems like a red herring. Reddit is proof that there's no technical reason why a community news site can't be run on a fraction of the cost of Digg.
The real lead here is that Digg is ceding that market entirely to Reddit. The new Digg looks more like a competitor to the Huffington Post. To be honest, I think there might just be a business there. Huffpo serves a particular market niche: predominately liberal female city dwellers and it strikes me that it may be profitable to deliver the same product to a different niche.
I think it looks very clean, very nice. I also think making it a place to find interesting content to share in people's social networks, were the majority of conversations are happening, instead of making people comment in-site, was a smart move.
94 comments
[ 2.9 ms ] story [ 90.5 ms ] threadhas anyone some details to expand on it?
When it launched, they spent the majority of the time attempting to scale it in a quick fashion - a process which probably involved buying lots of resources and attempting to scale as fast as possible. It was badly designed from the start, and was never really fixed.
EDIT: Actually, it looks like TC has a history of believing Cassandra = Bad for digg.com: http://techcrunch.com/2010/09/07/digg-struggles-vp-engineeri...
Yes, I forgot about Netflix! Not only do they use it, but they've provided tons of excellent tools for it (Astyanax, Priam) and blog about it very frequently.
Hence their previous statements. Kevin's statement was in reference to them switching completely to NoSQL and not knowing how to scale it properly.
Second, am I missing the comments on stories somewhere or is it just a website where links get voted on, but no discussion occurs?
more on their blog http://blog.digg.com/post/28441399381/welcome-to-digg-v1
2) Discussion on digg was always a bit janky. HN really spoils the idea of social links/comments because they are so high quality here, but when you're targeting the average person, its harder to control quality. Let people sound of on Twitter/FB, and you've created a new marketing vector without directly polluting the experience on your site.
That's an overstep that will keep me off the site.
My Reddit id on the other hand does not directly tie to my name, and that's just how I like it.
It's still fairly easy to find - I don't care if people who really want to figure it out, as I don't write anything I can't stand for. If I wanted true anonymity I'd take a lot of extra precautions.
I just care that my Reddit comments don't show up on the first few pages of a Google search for my name, as it lets me not think as much about my "professional image".
Putting it on Facebook is something I'd never let happen.
There's a huge difference between being possible to find what I've written if you really care about it vs. it being showed in the face of my family, for example.
And this is without being part of any controversial sub-reddits. There are plenty of sub-reddits or individual threads that are offensive enough or controversial enough that I'd imagine they'd die pretty quickly if people were forced connect their Reddit identities to anything that might be directly matched to their names.
My lawd man, build your own. You restricted yourself to 6 weeks, not users.
>Using Facebook for account registration is a short-term solution that will seriously cut down on spam, while we take our time to develop more robust spam-filter technology. We know this isn’t ideal, so rest assured: we are working towards a more lasting solution. [2]
1. http://blog.digg.com/post/28441399381/welcome-to-digg-v1#com...
2. http://digg.com/faq#why-facebook
"A lot" is a very relative term. Non-Facebook users are probably in a minority in the kind of audience they're targeting. I think it's an interesting choice. I'm not going to say it's sensible just yet, but when they introduce comments it will be interesting to see if forcing real names brings the level of discussion up.
"Using Facebook for account registration is a short-term solution that will seriously cut down on spam, while we take our time to develop more robust spam-filter technology. We know this isn’t ideal, so rest assured: we are working towards a more lasting solution."
Is anyone else finding this new redesign even more irrelevant than ever? What does digg provide at all now? No discussion and a very messy hard way to even find anything.
Would anyone consider this a ready to launch replacement? If they don't value commenting as one the of the most core features what hope do they really have?
To me HN is more valuable for the comments then the aggregation. I often skip the article and just look at the comments. Digg cannot now compete with this and if this is their priorities: big gaudy pictures, voting, and distant third or less, commenting... well.
good luck I guess but we have different priorities on what we were hoping for and expecting and need.
This is an alpha, or a beta, or whatever you want to call it.
Content-wise, it's not the same ol' garbage that keeps getting regurgitated on Reddit, or the same ol' subjects that keep getting rehashed to death on "Hacker"News. ("PHP is a terrible language!" "No it's not!" "Yes it is!" ... yuk.)
Visually, it's not a list of links which can only draw your attention by gaming the headlines. It's about time someone re-remembered that a picture is, indeed, worth a thousand words.
As far as comments go, it's been a long time since I've been consistently enriched by reading the comments on anything anywhere. (This thread is a good example, my blathering included.) I've got my fingers crossed for Web 3.5: "Here's a nice page, shut up and enjoy your stay."
My biggest complaint is the layout that seems to assume a huge screen or something wider than my laptop has. Scrolling left-right in addition to top-down is annoying enough for me to bounce.
I recognized most of the stories from Reddit.
Digg was a pretty awesome community in 2005 and 2006. The content was largely tech related and the comments were informative and interesting.
In 2007 it started getting worse, there were more jokes and memes and the content started to suffer. With no subreddit like system the original community began to become drowned out.
Towards the end of 2008 there was nothing left except memes and lame one line jokes that put youtube comments to shame. I was always surprised by all the people complaining about Digg v4 - the community sucked a long time before that.
I'm impressed by this revision, should be interesting to see what happens.
It's no secret that the community needs a reboot, and content-first development might work for a site that through any other implementation might just strike people as a poor version of Reddit (regardless of whether or not that may be true).
The team behind this is clearly very skilled. Will be keeping an eye on this as it evolves.
It also makes a ton of sense to have the algorithm factor in tweets, FB shares, and diggs. Nice. I don't know if it'll ultimately work, but it looks like a good start.
But, as always, such sites' value comes more from their community than anything else.
I was sure that the level of negativity on the company leading up to the IPO would pretty much assure it's success.
But I think folks have been proven to be mostly right about that (if not a little hyperbolically fatalistic).
but then "De mortuis nil nisi bonum" (Speak no ill of the dead)
It might be their version 1, but calling it "v1" in public makes it feel like the old Soviet pictures where the leaders who had fallen out of favor were airbrushed out (http://en.wikipedia.org/wiki/Censorship_of_images_in_the_Sov...). "Nyet, Comrade! Commissar Rose was never in charge of this tractor plant!"
It just feels really disrespectful to the original Digg team, at least to me.
The style itself is really nice, blocky and beautifully simple. This is how you redesign a content site, its elements are purposely made minimal, simple, and sterile so that the content stands out above everything else. The site disappears and the content takes the stage. Basically, it puts the spotlight on the performer rather than the stage. Also, the site's so lightweight that it just feels fast. Very few elements are begging for your eye's attention which is great, it allows the eyes to go straight for the image and title. Everything about the redesign is just so right, it's a shame they went for the magazine style layout.
Either way, great redesign. It's won me over, and I will actually start returning to digg to see how things go over the next few months. This was a very nice rebirth, back to simple, back to content.
Zite does this extremely well. Pulp, on the other hand, does it pretty poorly. Doing this on Digg's homepage is going to be challenging to not look schizophrenic. But on Digg's sub-topics, this type of layout will work well. I find scanning through tabular data to be fatiguing and I often scan over items out of boredom.
It's leisure information, not work.
This layout for digg fits in with casual consumption and discovery, rather than a structure that encourages seeking out specific articles.
I do agree with you that this is a beautifully simple site and will also revisit on a regular basis to see how it progresses.
Anyone know the tech stack they're using? Don't see anything specific on their website (blog, about, careers).
The real lead here is that Digg is ceding that market entirely to Reddit. The new Digg looks more like a competitor to the Huffington Post. To be honest, I think there might just be a business there. Huffpo serves a particular market niche: predominately liberal female city dwellers and it strikes me that it may be profitable to deliver the same product to a different niche.