Hacker News Reader

Comment by rkudeshi | Hacker News Reader

I wonder how much this has to do with the rise of Pocket over the last year.

It's no secret that Pocket has improved dramatically since rebranding a year ago. And I'm sure most people who are looking for their first "read-later" app are trying Pocket (free) before buying Instapaper ($5).

Personally speaking, I bought Instapaper for myself and gifted it to a few other people over the years, but I switched to Pocket a few months ago. The continuous updates and simpler interface (those thumbnail pics are surprisingly useful) won me over, despite my hesitations about their business model.

View on HN

Instapaper is nice because I felt like I was paying a reasonable amount for an app that I use regularly.

Pocket, although having its own benefits, is free. Any service with ongoing costs (parsing servers) that charges nothing for their app has me questioning their longevity.

I don't understand why people care so much about the "longevity" of an app that at most saves their data for a week, like Pocket, or something like Google's Keep, a notes app.

Who cares if it dies in 2 years? You might be switching through 3 other alternatives by then anyway. And you can still save your week worth of data if they announce the shut down anyway.

It's not like say an e-mail service, where you might actually want to keep your years worth of data.

>I don't understand why people care so much about the "longevity" of an app that at most saves their data for a week, like Pocket, or something like Google's Keep, a notes app.

I have thousands of items in Pocket going back a year and a half.

And I have hundreds of "important" bookmarks in Chrome that I'll probably never check again.

Let's face it, if you aren't reading it in the first week, you probably never will - except maybe that one article you plan on reading when you're travelling a few weeks from now. But even then you might forget all about it, with all the news coming it.

>And I have hundreds of "important" bookmarks in Chrome that I'll probably never check again.

That's a pretty terrible analogy.

What I have in Pocket is more than a bunch of bookmarks, it's a local archive of everything I've added. I can read it in a nice flowing text format and benefit from limited search functionality. I can reference anything in there anytime whether I have connectivity or not, whether the source page still exists or not.

>Let's face it, if you aren't reading it in the first week, you probably never will - except maybe that one article you plan on reading when you're travelling a few weeks from now.

Stop projecting your habits and use case onto everyone else.

if you have unread items going back for more than a year, chances are you are adding more to the list than you are removing. I guess hes right, i have observed this in myself and many colleagues from the days of delicious onwards. You bookmark/tag alot of interesting stuff because you are absolutely sure you will need it again and then forgetting about it after several weeks/months never looking at it again. In the case that you need that knowledge again, you just dig it up on google (if you can remember it).

Might not be true for you, but i am sure its pretty common behaviour. I have given up on read-later apps because stuff just kept piling up. Noawadays i just leave the tabs open until i have read something of interest and if i figure that it isnt interesting enough after a few days i just close the tab and forget about it. That way i am managing my to-read list on a daily basis and it doesnt go back later than a week. With chrome sync its also easy to grab the open tabs on my other devices and read them there.

That being said things might be different if i had a long communte or had more freetime after work, but in that case i am usually reading a book, if i am reading anything at all. Giving the brain some relaxation time from the constant bombardment of tech "news" is nice as well.

>If you have unread items going back for more than a year, chances are you are adding more to the list than you are removing.

I think you're missing the point and stuck in the use case of these tools as simple bookmarks or reading lists.

I have little reason to remove anything, ever. It's all being stored for free/negligible cost. I can reference it whenever and wherever I like.

>In the case that you need that knowledge again, you just dig it up on google (if you can remember it).

Right, I can search Google's index how ever many billion pages, hoping that what I want still exists online and that I can remember something specific enough about it to filter the specific needle I want out of the world's largest haystack.

Alternatively, I can just search the store described above which holds point in time copies of the relatively tiny subset of things I've actually seen or at least found interesting in the past. I can use utterly generic, half remember descriptions like "database" and get just a few matches rather than the 764,000,000 of Google.

>Might not be true for you, but i am sure its pretty common behaviour.

That's fine, I understand how people might choose to use these service differently than I do.

This tangent started with mtgx claiming an inability to understand why anyone would care about longevity or keep years of data in such tools.

I'm trying to help with that understanding.

I've done this in Instapaper. I have many thousands of articles (thanks to the instaright Firefox extension) going back years. Clearly, this is what Clay Shirky says is a filter problem, not a problem of too much information. I no longer horde in instapaper like I used to but I would like to at least be able to be able to export what I have since I started using it and even transfer the FULL archive to pocket. Probably doesn't apply to many, but: "If you have more than 2,000 articles, only the most recent 2,000 will be included. " I not only bought Instapaper, but I've been a paid subscriber since the beginning (so I can search my archive for potential research) and it would be nice to download a full archive, but nope, I can't.

I think OP's saying he has read archives going back 1 year and a half. Not unread, read.

I too have hundreds of read articles sitting in my Instapaper's archives, and a few dozens of favorites. I'm more concerned for them than the unread articles.

If Pocket goes away in the blink of an eye, what becomes of your archives & favorites?

>If Pocket goes away in the blink of an eye, what becomes of your archives & favorites?

Seems like the type of thing to be proactive about it.

Pocket has an API that you could use to sync with another similar service , or roll something of your own. They also have a simple export which provides a list of source URLs for what you've pocketed.

I've set Kippt up to retrieve from Pocket and used to the manual export to load things into historious [1].

As a last resort, on the Android version all the offline assets and pages can be found here [2].

1: https://news.ycombinator.com/item?id=5610492 2: com.ideashower.readitlater.pro/files/RIL_offline

I only read when I'm on holidays. I also keep warm clothes around even though I won't need them for another 6 months :) There are many ways to enjoy "slow IT".

That said, moving away from Instapaper was some grunt work but not too hard - I've opened every link in a tab and used hotkeys to add them to my new weapon of choice (Reading List).

This is absolutely right. We are drowning in an tidal wave of books/articles to read, videos to watch, podcasts to listen to etc. Trying to keep up is a fools' errand we must admit to sooner or later. The best we can do is give in and hope to drink a few sips from this ocean.

>Pocket, although having its own benefits, is free. Any service with ongoing costs (parsing servers) that charges nothing for their app has me questioning their longevity.

Prior to the name change, Pocket offered a $3 "Pro" version.

I'm guessing that Pocket means to be acquired eventually.

Pulse runs a free app parsing/serving from App Engine and took in $9.8M in funding before being acquired by LinkedIn for $90M.

Pocket runs a free app parsing/serving from EC2 and has taken in $7.5M in funding.

I was able to see who would be interested in Pulse and why pretty early on. I'm not sure about the same for Pocket.

I'm guessing that Pocket means to be acquired eventually.

And those acquisitions work out so well for the product, especially smaller, more niche products.

If they've taken that much money, they are going to have to provide a strategy to monetize. If they are doing it for free, it means they are selling analytics about what you are reading to somebody else.

For myself, I'll pay for the app.

Most paid for apps (though not all) sell exactly the same analytics that free apps do - why turn away an extra revenue stream?

Unless the developer has committed to not selling your data, they probably are already doing it.

Exactly.

I had invested time in Spool, and when they got acquired by Facebook, all I got was a bookmark list in a mail attachment and no more service. No readable article, no offline video, etc.

Even paying apps go for for a sale, less often though. For me Pocket is so seamlessly integrated with my online presence that it will of course to see it go(considering acquisition will kill it or somehow make it unusable - ad etc) but the ease/swiftness makes to very hard to switch to any other service paying or non-paying.

One example is their Android focus which, sometimes I funnily think, developers like Marco despise(no, read ignore :P ).

Instapaper's Android app, while not done by Arment himself, is probably the most pleasing non-Google Android app that I've used. He may not care about the platform, but he didn't push a poor Android product.

I subscribe to Instapaper but I switched to Pocket because I prefer their UI. I keep begging them to take my money.

You should be questioning how they make money instead. Because if it's free for you, you're probably the thing being sold in that scenario.

Indeed. Many companies are hesitant to fully activate their advertising apparatus, and go years without making revenues (eg. Tumblr). I worry when I see a company that isn't attempting to accrue meaningful revenues.

I'm actually quite happy when I see that companies begin serving ads substantially.

Speaking of Pocket, I'm baffled by their lack of (recently upgraded to simply lacking) sharing functionality and full-text search.

Passing a Pocket link to someone without a Pocket account leads to a sign-up page. If the recipient already has an account or goes to the trouble to sign up, they're redirected to their own queue, not the linked article. If the recipient goes to the trouble to sign up and pocket the same page themselves they will end up successfully accessing the getpocket.com/a/read/XXXXXXXXX URL that they were original sent.

Pocket is throwing away potential new users who quit at the sign up page and useful information about how existing users share.

Within the last month, they finally added some sharing functionality, but it sends links to the original not the pocket version. Another opportunity to pull new users into a test drive of the product being thrown away.

Lack of full-text search makes referencing anything down the road incredibly difficult. I've ended up regularly importing my Pocket queue to historious [1] to achieve this.

At this point, the only thing that keeps me using Pocket is the very nice Android app.

1: http://historio.us

We also do full-text search for Pro users, and you can sync with Pocket. http://kippt.com

I actually signed up for Kippt when initially searching for a Pocket alternative and there's a lot to like about it.

I didn't stick with it or sign up for pro for a couple of reasons.

Namely, lack of an app and what I find to be a "cluttered" experience.

Perhaps I'm missing something, but if I want to save the page I'm on (I know I can save a link directly with a right-click) to Kippt I hit the extension icon which prompts me for tags, a destination and requires an extra click to actually save. Contrast that with Pocket and historious where it's one click, done. Pocket actually does it best by temporarily presenting me with the option to add tags or go the list without insisting.

Now, if I want to search what I've saved on Kippt I head on over to kippt.com and I'm presented with something that (to me) resembles Outlook. I click into a little search field, type my search and hit enter. I wait, get results then click a link.

If the link came from Pocket, I get a largely empty page with yet another link to click to reach the actual content. (Links added directly to Kippt for later reading have a summary here, that's fine.)

Contrast that with historious. When I want to search what I've saved or imported from Pocket I head to historio.us and instantly it's Google for my stuff. I start typing and hit enter. My results appear immediately, if I want to go straight to source I hit the main link, if I want a reader view I hit mobilize below it, if I want to publish it on the public Google for my stuff at user.historio.us I hit publish.

As you can see, relative to that experience, Kippt requires "extra" steps for every common action I use.

Kippster, a 3rd party app, has been pretty great for me on iPad.

I was an avid Pocket user. And it's great for saving things you want to read later and then just throw away.

A lot of these articles/videos I want to read later, then save for reference. With Kippt I can send pages to my Inbox exactly like I did with Pocket. Once I read them, if I want to save them for later, I can move them into any 1 of the categories I've set up. I have about 20 categories ranging from "App Store Articles" to "Analytics Videos" to "General Programming Learning".

I can't imagine how I would live without it now.

> Passing a Pocket link to someone without a Pocket account leads to a sign-up page.

I'm not sure what you're talking about. I just tried sharing something from the Pocket Android app and it gave me the original URL in the Gmail message I opened.

>I'm not sure what you're talking about.

That probably has something to do with failing to read what I wrote.

I'm talking about a link to the "read later" view. I even described the original URL functionality you're confusing it with.

"Within the last month, they finally added some sharing functionality, but it sends links to the original not the pocket version."

Seems like it makes a lot of sense to do this. They don't want to become a de-facto mirror for the content, they want to provide it only to people who are actually their users. I'd rather have that, along with a bias towards sharing the original link, than a lot of sharing of the reflowed content.

>They don't want to become a de-facto mirror for the content, they want to provide it only to people who are actually their users.

Surely there's a way that could be prevented while retaining the benefit of exposing potential new users to the service?

The current sharing feature sends 800 character URLs pointing to email.getpocket.com. I'd think something could be put into and parsed out of that monster string to prevent the mirror scenario.

Send the first X clicks to a reflowed version with with a non-invasive sign-up option, send the rest to a sign-up page sans reflow.

>Passing a Pocket link to someone without a Pocket account leads to a sign-up page.

uhh… No it doesn't

>uhh… No it doesn't

Please read the whole post.

The analysis on his part had to be simple. He has two products right now, one of which generates $2.40 one time and then people use it for years. His second product pays him $1.40 per customer every month. The second product is far more stable long term.

I have no particular loyalty to Instapaper, but their major feature over Pocket, for me, is sending a compilation to kindle as an ebook (Kindle sees it as a magazine, complete with TOC.) People have been asking for this feature from Pocket for a while now (http://help.getpocket.com/customer/portal/questions/286894-k...), but without much encouragement that it will appear.

Here you go: http://www.crofflr.com/auth/login

Here you go: https://play.google.com/store/apps/details?id=com.tchami.kin...

I think Instapaper for iPhone is more polished and minimal compared to the new Pocket and that's why I like it. However, the web version is extremely trouble and useless and needs some work.

But when someone of my family or friends asked me to what should he use. I always says Pocket, because it is easy and not classy like Instapaper and don't forget the updates too.

It absolutely had something to do with it. I don't think anyone would say that Instapaper is 4x better than Pocket ($3.99 vs Free). If anything Pocket has clearly made a name for itself with better design and its new Pocket for Publishers API. I think Instapaper saw its peak about 6 month - 1 year ago and with The Magazine doing well, it was smart of Marco to sell it off in my opinion.

Instapaper's parsing is 3x better at least. Very often I see Pocket lose chunks of articles over innocuous divs.

And I'm often reading those articles in Pocket, which means I never see them at all. Very frustrating.

3.99$ vs 0$ means it is infinitely better not just 4x times better.

I don't think he was referring to the math involved (even though he wrote it) ;)

Yeah absolutely.

I tried Pocket and couldn't get past not having article length indicator, the quick archive/delete menu, favorites, and the Kindle auto-digest. Has the situation improved in any of those areas?

There is one simple thing that prevents me from using Pocket: pagination. It does’t work, it’s fiddly, it’s unclear how to turn it on, when you manage to activate it it switches off with the tiniest gesture. Pagination is the big Instapaper feature for me.

i could be wrong, but isn't pagination just swiping to the right to activate it? and then to deactivate it, you would just scroll the content normally.

that work for you?

The thing which irritates me most about this pagination model where you swipe, is that if there is a code reference or something off page, it means u wrestle with this page flipping on/off until you finally are able to drag the page so u can see remainder of code listing.

Not well enough.

Reply on news.ycombinator.com