Facebook's feed overflow fix will have collateral damage

Facebook, the everyman feed reader, has an “attention economy” problem that’s familiar to anyone who has used industrial feed readers like Feedbin or Reeder.app (emphases mine):

Facebook Will Curtail Unpaid Ads by Brands - Vindu Goel NYTimes.com

…So many posts, videos and images are being published on Facebook that the average user has about 1,500 new items they could see when they log on. Some people have as many as 15,000, the company says.

Over the last two years, the social network has repeatedly tweaked the system to show the top 300 or so items that it predicts each person will want to read. Facebook argues that people prefer to see videos, photos, news articles and updates from their friends and family more than from brands. So over time, posts by businesses have shown up less frequently….

The problem is Facebook users liberally “Like” all kinds of post sources - Pages (businesses, orbs), Friends, and Groups. Their Feeds overflow.

My son is up to several hundred sources, and they each generate a few posts a day. I’m sure he’s far behind the stream — if his timeline showed “most recent” he’d never see anything from Friends or Family.

Of course users could turn off the pages they don’t want, but, really, they kind of like them. The problem is compounded by Facebook’s UI; unlike an industrial Feed Reader (Feedbin, Feedly, etc) it’s not designed for managing large volume post streams.

So users are drowning in posts, most of which they kind of like, but they’re not seeing the things they really like — like the grand-dogs new teeth or the niece’s haircut.

At the same time Facebook is watching ad revenue go missing. Users dislike Facebook’s ads, but they like news posts from their favorite bookstore — and the bookstore’s paying nothing (or used to, anyway).

That problem is getting fixed …

… the company told marketers that if they wanted to reach customers on Facebook, they needed to buy an ad.

…  fewer fans of a retailer will see its notice about a big sale and fewer fans of a video game company will see a post promoting its latest app.

Even posts from big advertisers that spend millions of dollars on Facebook ads will vanish from the news feeds of their fans unless they turn them into ads…

Seems reasonable. I know how to tweak my Facebook stream to see what I like — but I’m a geek. Civilians aren’t going to do that. So the “fix” doesn’t seem too bad — unless you know something Mr. Goel doesn’t know. Unless you know that not all Facebook Pages are businesses or “brands”.

For example, these are 6 Pages I administer:

Looking at my own Pages Feed I see a novelist friend, the Major Taylor Bicycling Club of Minnesota, Saint Paul Bicycle Coalition, Keweenaw Trails, Cyclocarbon, Avalon Charter School, CrossFit St Paul, Minocqua Winter Park, Autistic License Play, Lebanon Hills Mountain Bike Trail, City of St Paul _ Government, Skate the Oval and a bunch more like that. A couple are businesses, but mostly it’s non-profits and local government.

All of these are at risk - and the trend is already underway. Here are some recent stats from our special hockey team page - a page with 81 “Likes” (Subscribers):

 

The most recent post was seen by 15 of those 81. That number is going to fall even further — unless I pay for the posts to be seen (I won’t).

Facebook has a real problem — but their Feed Overflow Fix is going to cause a lot of collateral damage. 

One approach for organizations and nonprofits would be to convert Pages to Groups - except we don’t know how Facebook is going to treat posts from Groups. Maybe they’ll want Groups to Pay to Play too.

For my part I’ll delete most of these Pages, and I’ll move back to email and to blogs. The transition is easier because so many of the people I want to reach have left Facebook (perhaps because of the Feed Overflow problem).

It’s not all bad news. Facebook’s policy might move non-profits and local government back to open-standards blogs - eventually.

A variant of English optimized for machine translation to Chinese

A Jonathon Duerig comment on App.net reminded me of an old idea that still hasn't come to pass.

It should be possible to define a style of English expression, including vocabulary and grammar, that lends itself to machine translation, particularly translation to Chinese. It would likely resemble the language we are learning to use with Siri. We can call it TransEnglish.

With the technology we now use to correct grammar and spelling, and predict text entry, we could do dynamic rewriting -- identifying or rewriting ambiguous phrases, substituting more specific words, revising complex tense structures.

A test harness would machine translate to Chinese then back again, then identify and close divergences. A nice exercise for neural network tools.

I expect after a few weeks of use an English writer would quickly learn how to write translatable text naturally.

Similarly, I'd love to follow some Chinese authored blogs that were written in a similar machine-translation friendly flavor of Chinese - TransChinese. (Indeed, the resulting English output might well be of the same form as TransEnglish.)

AT&T secret mobile opt out system shows sense of humor

Nick Heer described AT&T’s customer activity tracking best …

AT&T Stops Using Undeletable Phone Tracking IDs — Pixel Envy

… Here we have a case of AT&T actually doing the right thing. They get criticized so frequently for so many reasons, so I think it’s important to point out when they do something good and ri—

*mimes touching earpiece*

What’s that? Oh…

… Dicks.

Briefly, both AT&T and Verizon track the sites we visit so they can serve appropriate STD treatment ads based on Dad’s net browsing. Verizon still does this, but AT&T claimed to have reformed. Except they really haven’t…

Edmonds said AT&T may still launch a program to sell data collected by its tracking number, but that if and when it does, "customers will be able to opt out of the ad program and not have the numeric code inserted on their device."

Really, I kind of prefer Verizon’s honesty. Being a Verizon customer is like joining Sauron’s army — there are no hurt feelings when your commander rips your throat out. Because that’s just what Verizon does.

AT&T does have a great sense of humor though — here’s the opt out link you’re supposed to visit while on AT&T’s network:

AT&T Adworks - http://205.234.28.93/mobileoptout/

Yeah, an IP address. I think it’s legit, fwiw the footer says “AT&T’s intellectual property” …

Or maybe it’s just a malware infested Serbian honeypot.

Or a psychology experiment.

Or performance art.

AT&T, you kill me.

PS. I suspect AT&T resellers (MVNOs) get the same treatment.

After the Apple Watch debacle - the Nano recovery

Seven years ago Clayton “Innovators Dilemma” Christensen wrote ..

Clayton "innovators dilemma" Christensen: Apple will fail

… the prediction of the theory would be that Apple won’t succeed with the iPhone. They've launched an innovation that the existing players in the industry are heavily motivated to beat: It's not [truly] disruptive. History speaks pretty loudly on that, that the probability of success is going to be limited…

By “existing players” he meant Nokia (now a forgotten part of Microsoft). That’s the problem with making testable predictions — they break theories.

Which brings me to the aWatch, of which I am not a fan …

Gordon's Notes: Apple Watch - a bridge too far

… I don’t think the 1st generation Apple Watch will be nearly as successful in the US market, though it may have some success in its true target market of China. Unlike the much loved Nano-clip it doesn't solve anyone's problems well. An water-susceptible exercise device tied to an iPhone is far less useful than an inexpensive FitBit. An authentication device tied to an iPhone is redundant in today's world. The iWatch Apple Watch is a very limited music and video platform. It’s too big, it’s too expensive, it's too fragile (water), the battery is too small and the initial demo highlighted bumping hearts...

… A waterproof $150 iOS 8 Nano-clip replacement in Sept 2015 will be interesting. Splitting the cellular phone into multiple components, for which iPad and Apple Watch are interaction elements will be interesting. Standalone Apple Watch 4 running on next-generation LTE will be interesting.

Apple Watch 1 is a mistake.

The aWatch will launch in the US and Chinese markets in a few months. It will fail early in the US market. There will be initial success in China, then it will fall to China’s chaotic nationalism and less expensive and more useful Chinese clone-variants. It may have some persistent sales in Japan.

So what happens after that?

Jonathon Ive either leaves Apple or tolerates a diminished role. He’s very wealthy and has accomplished much, so we shouldn’t feel too sad for him. Tim Cook moves his executive team around and puts his rhinoceros skin to good use. The share price dips and returns to trend line.

I think that will all be good.

The interesting bit is what happens to the aWatch tech and how soon will we see it in another form?

The timing depends on what Apple really thinks is going to happen to the aWatch. I assume that some execs expect it to fail and that there’s a plan B, and maybe a Plan C, in the works. So what should we expect in the fall of 2015?

Physically the Plan B device looks a lot like the much beloved 6th Generation Nano Clip. It will be designed to work with a wrist band or a clothing clip. It will be an excellent 32GB store music device but will also act as a detached extension companion to an iPhone. It will be good at caching data and then posting it back when in phone range. It will have some GPS functionality (limited) and some exercise tracking ability when attached to the wrist.

Plan B will be modestly successful worldwide.

Then there’s Plan C. I owe Plan C to @duerig, who carries an ultra-slim flip phone and an iPad. I’m convinced he’s got things right — that the world is going to go towards people who carry just a phablet and people who carry a phablet and a mini-phone. Apple’s got the phablet market covered with the 6+. Plan C is a slightly larger and heavier version of the 7th generation nano paired with an iPad Mini 4 [1]. This iPhoneMini is a device Apple considered launching in 2011 and the iPad Mini 4 replaces the already forgotten iPad Mini 3 (Apple’s feeblest product hop ever).

I might buy Plan B (iOS nanoClip), I would definitely buy Plan C (iPhoneMini + iPad mini 4). 

Both of these are good futures that will leverage aWatch investments. Look for Cook to announce them when Apple buries the 1st generation aWatch. Which means that stock dip will be short-lived.

Keystone question: if US were to meet China accord goals, would Keystone be economically viable?

To widespread surprise, the US and China reached an agreement on carbon emissions:

U.S. and China Reach Climate Accord After Months of Talks - NYTimes.com

… United States would emit 26 percent to 28 percent less carbon in 2025 than it did in 2005 …

Which brings up a question that seems obvious, but also goes unasked.

If the US were to meet this accord, we’d have reduced carbon emissions by some combination of Pigovian taxation, regulation and technological innovations. However we achieve that end, wouldn’t this reduction make Keystone and similar projects economically unviable?

If so, then the Keystone project is a bet that the US will fail to meet this goal. Further, it will be a powerful sunk cost incentive to ensure that we fail.

If Keystone’s business case makes sense in a world where we reduce carbon emissions by 28% relative to our 2005 baseline then build it.

Otherwise, don’t.

I don’t think this is very complicated.

Annals of return to RSS - Geeks who don't realize their site has a feed

RSS, left for dead years ago, has crawled back out of the wilderness. Today we welcome our latest Prodigal - Jon Udell.

There's still work to be done though. Consider Data Elixir, data sciences site based on the "Curated" email platform. True, big data isn't cutting edge any more, but this is still a relatively modern site. So modern it supports two ways to subscribe - Email and "Safari Push Notifications".

No RSS.

Except, of course, the site does have an RSS feed. Works fine. As best I can tell the authors have no idea it exists.

This will take a while...

Election 2014: post-mortem

Why did it turn out the way it did?

• electoral map
• midterm bias to GOP base (turnout)
• voters unhappy with personal economic experience.
• reversion to mean on social issues - especially male response to gay marriage, women's issues

Whose interests are served by outcome?

• the 1%, especially the 0.1%
• megacorps (corporate tax reduction)

What policies are likely to be advanced?

• corporate tax reduction

What policies will now be delayed or reversed?

• ObamaCare (McConnell means what he says)
• Carbon control
• Immigration changes
• Economic equality agenda
• Bank and finance controls
• Disinflation response

What are implications for 2016?

• Probably helps Dems

What do I think?

• Disappointing, not surprising.

She's dead Jim. Requiem for a 9 year old G5 iMac

I bought my G5 20” iMac in July of 2005 with a 400GB SATA drive and SuperDrive for $1879, it expired today at 9y3m of age. I think the 400 GB Hitachi Deskstar is original - so it is the longest lived hard drive I’ve owned. I pulled it and was able to read it from a drive cradle, though there’s nothing left on there we need.

Today the comparable machine in Apple’s lineup is the 27” 3.2GHz iMac with a 1TB SATA drive but no SuperDrive for $1799. Of course that’s a far faster machine and a much nicer display, but the price is remarkably similar. Not to mention that we’d need to add an ugly external DVD player.

I doubt the 2014 iMac will last 9 years, not least because it’s costly to service. The G5 iMac was designed to be user serviceable — perhaps because Apple stores weren’t ubiquitous. The back came off with 3 standard Philips head screws. It was trivial to swap out a hard drive or memory. When the power supply died in 2011 ($150 for iFixit, there was a recall @2006) it wasn’t hard to swap it out. I think that replacement power supply just died; this time I decided a repair wasn’t worth the trouble.

The G5 iMac was not trouble free. Besides the power supply issues many died young from capacitor failure (defective manufacturing) — though I somehow missed that. The G5 ran very hot, it took Apple a year or two to figure out a reasonable fan management solution. Worst of all, the display developed internal discoloration [1]. 

Despite the issues, 9 years is a pretty respectable lifespan for a heavily used computer. We got a lot of use out of that machine, not least as a handy DVD screen.

Not a perfect machine by any means, but not bad.

- fn -

[1] My 2009 27” iMac has a similar discoloration though it fades with use. It has had two drive failures in 5 years, the last time I put in an SSD.

See also

• What do do with an old iMac that shuts down unexpectedly?
• iMac G5 rev 2: a lot cooler? 6/2005: the first gen iMac G5 was a train wreck. I really try to avoid 1st Gen Apple products.
• Fix for 10.4.3 fan problems: unplug and restart and resetting the SMU (PMU?)
• My iMac screen - dark band at the top of the display

The relativistic version of the linear motion equation: velocity = acceleration*time.

My son is starting to do … physics.

It is fair to say I am more excited about this than he is. There is a reason I have a copy of the 1973 Edition of Misner, Thorne and Wheeler’s Gravitation ($35!) on my bookcase. I’m saving it for retirement, by which time Desmos.com will breeze through all the visualizations. It is true I was a failure (non-physicist) at Caltech, but there’s still time…

Ahem. So when #w was given an assignment to create a poster about linear motion, with equations on time, velocity, constant acceleration and the like, I thought it would be fun to show that the equations of linear motion are merely a low velocity approximation to the relativistic equations of linear motion. I was sure a quick Google search would turn up a simple simulation that would look like:

v=a*t (starting at rest velocity after time t is acceleration times t)

Riiiight. I’m sure that simulation exists, but I was never able to find it. With a bit of thought about the problem I realized I might be better off dropping the tricksy concept of acceleration and looking for a relativistic version of:

v= (F/m)*t (since change in velocity is Force*Time/Mass. Push a trike, push a truck, which moves faster?)

That found an “off-topic” [1] stack overflow article which was just a few parentheses short of the good-enough equation [2]: v = c * tanh(asinh((F*t)/(m*c))).

This is the first time I’ve personally run into hyperbolic trig functions, so I thought that was pretty cool, especially since I’d just read Jon Butterworth’s lovely description of how one hops from simple physics to quantum physics simply by tossing the square root of -1 (i) into a classical wave equation (uses Euler’s Formula, so extra points). That article used the familiar sin/cos functions, so I was getting an extra dose of trig. 

Plugging v = c * tanh(asinh((F*t)/(m*c))) into Wolfram Alpha gave me this cool output (yes, the AIs wll be our death, but for now they’re fun):

Now that is what I was looking for! It clearly reduces to v = (F/m)*t when the squared stuff is relatively small. [5]

We can now compare my son’s high school textbook linear motion equation (v=F/m*t) to the relativistic equation, using the sneaky physics trick of geometrized units so c=1. Just to make things even nicer I’ll arbitrarily set the applied force to “1” (some unit) and the mass to “1” (some unit). This is what Desmos shows:

 

The last shows how simple this is in geometrized units (v is change in velocity over time t in our funky units):

and here’s the magic graph that shows what happens as velocity (y axis) approaches “1” (speed of light) in the original linear motion equation and the relativistic version:

Yeah, hits the speed limit. [5]

—

[1] The iron law of StackOverflow is that any article of interest will sooner or later by marked off-topic.

[2] I discovered the problem when I plugged it into Wolfram alpha, I had to do some web searches and play around the parens to get the correct expression.

[3] Incidentally, if you’d told me in 1995 that we would still lack easy entry of mathematical notation into web pages in 2014 I’d have assumed some kind of worldwide civilizational collapse.

[4] Once I had this expression, which was provided by the Wolfram AI, I found a discussion thread (scroll down) telling me it can be derived by “substituting F/m for a in the Baez equations(where m is the proper mass) and dividing by c where the tanh/sinh stuff is related to “proper and coordinate time”.

[5] I’ve done great violence here to the principles of general relativity — the meaning of time and distance are entirely dependent on frames of reference and I’ve glossed over all of that — largely because, you know, I’m not a physicist. It is kind of neat, however, what one can kludge together with a handful of web tools. In the general vein of non-physicist at play, it’s fun to compare this to Butterworth’s masterly representation of quantum physics using high school math. I again am left with feeling that physics will be easier to understand once we figure out what “distance” and “time” emerge from.

Blogging is definitively back - the NYT has redone their Zombie RSS page

My feeds (Feedbin/Reeder) have never gone quiet — but there’s no doubt it was getting harder to find links to feeds over the two years since Google tried to kill RSS to boost G+.

Then came signs of a turnaround. Apple very quietly added RSS reading back into Safari — after removing RSS from both Safari and Mail.app [2]. Google, fairly quietly, backed away from G+ — I don’t get any G+ social invites at all any more. More interestingly, Google blogs all became more active. Microsoft kept RSS features in IE 11. Facebook never removed RSS from Pages. Old blogs started lighting up in my feed reader. Rosenberg has started writing about a blog revival amidst disaffection with Twitter and Facebook [1]. 

All significant developments, but they pale next to the very biggest sign of them all — the New York Times has updated their RSS - Feed Page! It no longer recommends use of Google Reader! [3] The NYT has even added Topic Feeds:

Times Topics feeds collect news, reference, photos, graphics, audio and video on thousands of subjects, covering material published since 1981 … Search 10,000+ Times Topics Feeds

 Dave Winer should be a happy guy today.

[1] Not directly related, but fairly suddenly, and for no obvious reason, many of my friends and family have stopped posting on Facebook.

[2] Apple needs to update it’s RSS Feed page though — it doesn’t mention use of Safari.

[3] It does mention AOL Reader. I thought that was a bad sign, but, and this shocks me, AOL really does have a Feed Reader with its own spiffy web site: "Moving from another RSS reader? You can upload your subscriptions in standard OPML format and start reading right away!”. It even has its own friggin’ API. Turns out this was launched a year ago. It’s still in beta, but there’s an active development blog and they released an iOS app in August that has few ratings but seems well liked. Best of all, it supports OPML export as well as import. So this is a real contender.

The Alibaba/Apple Ratio and the fools guide to post-bubble market timing

Google tells me Apple’s market cap today is 576.39 billion dollars. (Think on that for a moment.)

Alibaba’s market cap is 223.28 billion dollars.

So the Alibaba/Apple ratio is 0.39.

I’ll be buying into the S&P as the market falls, but I’ll buy more when that ratio gets closer to 0.10.

Repairing carbon frame mountain bikes: Ruckus Composites vs. Cyclocarbon

With this post I'll finish my 3 week journey from carbon frame naivete to bloodied veteran (see below for related posts). It began when a bike shop refused to repair my newly acquired used Cannnodale Scalpel because of this frame finding:

Yes, the strange irregular line where the end of an aluminum seat insert meets a carbon frame top tube. The bike shop and everyone else who looked at it was certain the frame was dangerously unride-able, whereas the previous owner felt it was fine. (As it turned out, it's not clear who was right [2], only that there was no way to service the bike.)

After a bit of research guided by local bike shops and mountain biking friends, I learned about the hierarchy of carbon frame repairs. There are roughly 3 levels of carbon frame repair:

• Elite: often with a national market, may use ultrasound to inspect frame, usually need to strip bike and ship frame, repairs take 4-6 weeks but look like new.
• Regional "race ready": services a metro area or geographic region, mostly hand delivered bikes, focuses on fast turnaround rather than fine finish. Often used by racers, who own the majority of broken frames.
• Hobbyists who do a small number of frames, often for friends.

I ended up comparing an elite option from Ruckus Composites with a "race-ready" option from Drew at Cyclocarbon. This is what I heard back from Ruckus after they reviewed my blog post:

Ruckus Composites 

... First we will do a full inspection for structural integrity on the frame and hunt out all of the damage. once that is complete we will contact you with a quote and report our findings. 

After that we’ll machine out the damaged carbon fiber because (like a crack in a windshield or a tear in your jeans) all of the damage needs to be removed or it’ll keep propagating under the patch. 

Once we have that taken care of, we can go in and structurally rebuild it with fresh carbon fiber. When that’s done, we can repaint and clearcoat your frame so it’ll be just like new. 

For a repair like this I would estimate around $400-500 and about 3 weeks of in-house labor to complete. I suggest the next step you take is to take your bike to your local friendly bike shop and have them tear it down and box it up. Tearing the bike down to just a frame makes shipping a lot cheaper...

Cyclocarbon (Cyclocarbon on Facebook is more active) doesn't do this "like new" kind of repair. They do "race ready" for $150-$200, typically in about 3-5 days. That means no repainting of the original decals, no clearcoat, and a functional rather than "like new" shape. Repair warranty is informal.

I ended up going with Cyclocarbon, in large part because of turnaround time and because Emily was willing to drive my bike 1.5 hours from St Paul to Rochester MN [1]. Cyclocarbon has a strong local reputation, so I felt their repair would be reliable. For me the cost and hassle of tear down, shipping, and reassembly outweighed the aesthetics of a Ruckus repair. Once I factored in shipping and assembly costs Cyclocarbon was also significantly less expensive.

I think Ruckus would have been an excellent choice too however. I do miss the Scalpel graphics, and my top tube now has a bit of pudge. If I'd gone the Ruckus route I'd have invested in tools to do my own disassembly and reassembly.

It's good to have choices.

[1] If she'd known who lousy the drive would be that day I'm sure I'd have done local shipping rather than driving. I was able to pickup the bike during a trip to Rochester for my son's High School mountain biking race.

[2] Cyclocarbon's Drew tells me that despite the worrying artifact the carbon seemed strong. He's not sure whether the frame was truly weakened.

See also:

• What is the best material for a mountain bike frame?
• How to buy a used bicycle - and why carbon frames are a special problem
• Information asymmetry and can my Cannondale Scalpel purchase be salvaged?

PhotoCard 1914

I'm a fan of Bill Atkinson's PhotoCard.app;  I send one weekly to my father, currently in long-term care. I love the fusion of ancient Egyption and modern American tech.

So I appreciated discovering a 100 year old PhotoCard in my father's effects. This print is a (way) pre-photoshop overlay of my father's Uncle Max and his mates on an 'ISOLATED' image.

On the back of the print is the stamp that made it a 1914 PhotoCard - with a section for postage and for a note. (In this case it was never stamped).

Hard to believe there will still be postal delivery in 100 years, but maybe Amazon will stick them in packages.

What is the best material for a mountain bike frame?

One of the great things about blogging is that it lets someone who knows very little look much the same as a world class expert. All of the usual cues to merit and expertise are stripped way. In the Feed Reader, nobody knows you're a naif.

Ok, that's enough informed consent. I'm now going to tell you what material works best for a mountain bike frame - Aluminum, Steel, or Carbon. I base this wisdom on two weeks of recrimination after Akerloff's information asymmetry and rampant bike lust [1] got me a problematic Carbon Frame Cannondale racing mountain bike. During this time I sorted out how to buy a used mountain bike and why Carbon frame warranty loss hits resale value -- and I came up with the definitive guide to the BEST frame material:

Attribute	Aluminum	Steel	Carbon
Perform	2	1	4
Retail $ [2]	4	2	1
Repair $ [2]	1	1	3
Reliability	3	4	1
Comfort	1	2	4
Lifespan	2	4	1

The spider graph [3] makes it clear ...

What do you mean there's no clear winner?!

Year, it's like that. Since I don't race and don't have time to spend on fussy stuff I'd be best off with an aluminum frame or, if I had the money, a steel frame from one of the elite Minnesota steel bike companies. Still, there's no denying the performance and comfort of carbon.

So it just depends.

I lied. Sorry.

[1] My brother bought a Porsche under similar circumstances. He says I got off easy.
[2] In these cases higher rating is less cost. Carbon can be relatively cheap to repair if you can find the right person or don't mind mailing a frame.
[3] It is really quite amazing that one can create a table in Excel/Windows and paste it into Blogger's composer window as an HTML table. I don't think I can do that on my Mac. I tried to do this graph in Google Spreadsheet but didn't seem to be supported there.

Apple kills yet another photo sharing service - and generally screws up iOS photo management

I expect Apple to screw up anything related to long term data management, but this is extreme even by their standards. GigaOm, in language restrained by fear of Apple, tells us of another Apple datacide and botched product transition.

In the article quoted below “iPhoto” is iPhoto.app for iOS. Given the timid language, I’ve added some inline translation in square brackets …

https://gigaom.com/2014/09/27/where-are-my-photos-how-to-use-the-new-photo-management-features-in-ios-8/   

[iPhoto.app has been removed from iOS, replaced by Photos.app with fewer capabilities, including loss of iPhoto Web journals.] 

… In November of 2010 .Mac HomePages gave way to MobileMe Web Galleries. Then in June of 2012, MobileMe Web Galleries ceased to exist as iCloud came online. Now the most recent successor, iPhoto Web journals, is being shut down, or at least that is how it appears. With each transition, users of the previous online journaling feature really had little to no options available when it came to migration to a new or replacement feature. [users were totally screwed and lost hundreds of hours of work with no recourse] 

… you could add titles, insert comments, include maps, weather and other information intermingled with your photos. Users of journals would typically spend a good amount of time personalizing the delivery of their online photos by telling a story alongside their photos. 

The problem this time around is that there was very little notice and there really is no recourse or action that can be taken to preserve your iPhoto projects. … “Photo Books, Web Journals, and Slideshows are converted into regular albums in Photos. Text and layouts are not preserved.” And thats it, no more iCloud scrapbooking per Apple. 

… Apple has finally removed the concept of the Camera Roll …  all of the photos you have taken, whether they are on your device or not, now show up in the same “Recently Added” folder. This is not just a simple name change, it is a completely different experience. All of your photos are now synced across all of your devices, or at least the last thirty days worth. 

… iOS 8 has actually made it even harder to delete photos stored on your device [image capture delete all no longer works]. Tap and hold a photo in your “Recently Added” album and delete it from the album. It will move into the newly created “Recently Deleted” album …  delete it again from the “Recently Deleted” album…

Apple is a bit of a serial data killer -- usually with no public response. I still miss the comments I'd attached to iPhoto albums that were lost in the transition to Aperture.

Speaking of Aperture, both iPhoto (and Aperture) for Mac have been sunset, though Aperture is still sold. All three are eventually to be replaced by “Photos.app”, which may be an improvement on iPhoto but is certain to be a disaster for Aperture users. We can expect a large amount of personal metadata to be lost. (No, Lightroom is not a migration path.)

New users may be transiently better off once all the pieces are finally in place -- until the projects they invest in disappear. This is a cultural problem with Apple, not a bug that will get fixed. Never make Apple the owner of your data.