It Shouldn’t Take The ‘Crazy Ones’. It Should Be Everyone

This whole iPad isn’t thing.

I wish I saved the blog post I wrote back on my Wordpress account in 2012 about my previous computer purchase. I spent some time praising the joy of owning a retina MacBookPro and I remember typing, “this might be the last laptop I will own”.

FullSizeRender.jpg

This thinking came from all the power and versatility of the device I typed the review on at the time, it was spec’ed with the highest custom options available (about $28,000 HKD at the time). I felt I wouldn’t need another Apple laptop ever again. It felt that powerful in use.

Since that time however, Apple have tried to prove my now half decade old assertion correct. My 2012 purchase still feels greatly justified against the newest iterations of the rMBP. Everything after 2015 feels like a bigger than normal compromise one would want to make with an Apple product. Normally the balance in new features against the removal of the old ones works itself out. These newest devices haven’t had their “balancing” work itself out still and Apple won’t save face by adding back some legacy just to make it easier

Clearly Apple think they’re on the right path with their laptops. Who am I to argue? Seriously though, the MacBook Pro’s are a no-go for me

What else then could I upgrade to instead? I figured the iMac, (I know not a portable) actually was the next most appropriate because the price to performance ratio was much better and I’d only need one bloody dongle to move forward judging by the peripherals I own and the legacy ports still available to the iMac.

The October Event

The 2018 iMac update didn’t come as many presumed it would, Apple updated other we were crowing about and consequently the prices on those devices rose, just like with the MacBook Pro’s. This gave me reason to be concerned; that Apple is going to do this with the iMac line. Time will tell on that front.

The iPad was also updated and judging the product from the event’s marketing, it was supposed to be faster than any current laptop (around 92% to be exact) on the market and be a replacement for a laptop.

My laptop has been off the market since 2015 so I did some mental calculations that made an iPad Pro purchase seem like the logical choice in terms of its low price (relative to what’s available and what I paid six years ago for my main machine).

Ironically those “relatives” allowed me to sidestep the conversation people were having about the increased price of the new iPads because my personal circumstances gave me different criteria to compare against. The cost of computing for me, actually got cheaper, by a third.

33F2195F-EF06-45D0-9FBF-9B0C9C5E51F7.JPG

A 2012 MacBook Pro at $28,000 divided by 6 years of ownership equals to $4,666 per year. Using that as my baseline a $18,000 purchase (the 1TB iPad Pro* including pencil, stand, keyboard, 4TB WiFi hard drive) over 6 years would come out at $3,000 per year.

I think it’s reasonable, not just for me, but for anyone, it’s prudent to think longer term for computing purchase and it’s not out of the question to realise an iPad can be still be useful after 4 years. Heck the current iPad mini still sold by Apple hasn’t seen an update since September 2015. Apple seem to think so to. Again who am I to argue? :-)

I think I need to make another assertion :-p – “I can keep this iPad as my main machine for the next four years at least”.

The big conversation online however is whether an iPad can replace a laptop. This conversation was held previously for laptops replacing desktops and look what happened, nobody cares now.

At the moment people care again about going from laptop to tablet, while others clearly see the power of the alternative in an iPad.

What do I do on a laptop that CAN be done on an iPad?

Two things slowed my 6 year old laptop down, editing video with Final Cut Pro and browsing Reddit.

Those two things CAN and are being replaced as Jonathan Morrison recently illustrated to his large audience and Henny Tha Bizness has been showing how he works on an iPad since before this year’s conversation about whether creative work is possible on the iPad really kicked into gear.

Apart from some asides (Xcode), there really is no excuse or argument against not seeing a large subset of non-traditional (pilots, medical staff and sports tacticians) and traditional professionals (photographers, video and drawing artists) alike, using the iPad Pro in some sophisticated ways. People are proving it today, not later when iOS 13 supposedly will fix everything wrong with the iPad.

I’ve already edited my next YouTube video with an iPad and it was faster than I thought. Of course because the hardware is enabling me more than my laptop can but in addition, the capability of Luma Fusion in this case for manipulating said video is also making possible.

The desktop to laptop transition was an easy one because the UI paradigms and desktop metaphors kinda stayed the same across all the platforms during that time.

This next transition from laptop to tablet requires not only newer UI workflows, but pioneering, creative minds that say ‘yes’ in the face of a chorus of ‘no’ or ‘not yet’.

I’m saying yes, and my bank account agrees with me.

*As a general purchasing rule I believe buying the very most is the wisest thing to do when making an Apple purchase. It grants you longevity while using the computer and removes any performance doubts if you had skimped on an upgrade.

What Lies Ahead for Periscope - Discoverability (Part Two)

What lies ahead for Periscope - Monetisation. Discussed the issue of monetisation for creators; how they shouldn’t wait for Periscope to build in monetisation features. I advocated the idea to create their own revenue streams now, so as not to be solely dependant on Periscope. This came about because creators on YouTube had their channels demonetised. Those creators (on YouTube) that haven’t suffered, have other sources of revenue as a fallback. Periscopers should do the same. 

Part 2 deals with Discoverability, Part 3 deals with the quality of the live-streams present on the Periscope platform.

Discoverability

One of the largest concerns from the community since late 2015 has been one of search and discovery; finding the best broadcasters possible. The Periscope app hasn’t made any real progress towards finding consistent, and quality, live-streams in its two year history. 

One implementation that was seen as a step towards allowing viewers to simply find something to watch was the VIP program. The VIP program enables VIPs to give the Periscope editorial team a heads up on live-streams they think are worth promoting (a VIP can in fact notify Periscope and the editorial team will review it live in order to feature it).

Currently the most fruitful way (in terms of quantity) of finding a live-stream involves looking at the list view. The list view provides thumbnails and titles of broadcasts, all streams sit equally on the list (unless featured), much like your homepage on the YouTube app. The immediate difference though to the YouTube app; those videos on the YouTube app are a mixture of what the app thinks you may want to watch and from those creators you already subscribe to.

Pictured above are the four main screens with which to find broadcasts on Periscope vs the three screens from YouTube pictured below.

Aside from the list view on Periscope, you can find streams on the world map (streams are seen geographically). There is also the ‘TV’ tab, showing the recent broadcasts of those you follow. This aspect of the app does have some variation to it. If a stream is of importance, Periscope can push it to your device to ‘feature’ above the existing replays on that list and provide a category bubble further letting viewers understand the context, it may be ‘Travel’, ’Politics’, ‘Culture’ and so forth. These featured scopes depend on a 24 hour editorial team. 

This is short summary goes towards explaining what discovery currently is on Periscope. As there is no true search capability (category search) and understandably, looking for something live is quite a task due to the nature of the stream ending at any moment. Search has to lean on the broadcaster rather than the content. Finding something to watch on YouTube is no contest because it’s the largest video library in the world; with videos also having less ephemerality and tagging built in compared to broadcasts on Periscope.

How does discoverability get better? I’m aware of a Twitter endeavour to algorithmically determine the content of live-streams to make the identification process easier. That aside (which isn’t out yet) what is the solution? A similar intent is being made towards nudity.

Something that could have been implemented in 2016 was the ability to let the scoper decide on their own category bubble, from that, a basic form of categorisation could be collectively formed by broadcasters and used as another form of search by viewers. The implementation of this type of tagging and categorising towards broadcasts can also be organised in a summary based on the last five streams. 

If I made five streams with a tag like ‘travel’, then I’m in the travel section of a gallery list of other likeminded topics. I could automatically transfer over to politics if (because I’m in Hong Kong and Hong Kong people protest as a past time) the next three streams were tagged ‘Politics’ (three more streams puts me over the middle of five as the preceding last two would still be travel related).

This would also solve the existential issue broadcasters might have if they think they fit in more than one category. The app decides dynamically, balancing the broadcaster’s position based on the frequency of the tags . If the user forgets, the system ignores the scope, until the majority of untagged scopes takes you out of any category.

This way, viewers have an extra avenue in search functionality towards discovering other broadcasters and the proposal is simply extending the existing system used by the editorial team. It isn’t wholly a brand new form of discovery to design, develop and test for, for the Periscope team. The algorithm isn’t complex either, it simply sorts broadcasts on a weighting scale determined by broadcasters willing to tag their broadcasts.

The VIP Program and Search Results

The VIP Program is supposed to be the other way of discovering and boosting the visibility of Periscopers on the platform. In retrospect the VIP Program seem to be a solution that goes the long way round towards enhancing discoverability on the platform.

The VIP Program requires a lot of behind the scenes effort to implement and maintain by the Periscope team, its been in operation for quite some time now, with VIP scopers reflecting on its actual aims. I’m one of them and I’ve always thought the VIP system would come in phases, currently it still seems to be on phase one of an implementation, like it’s still unfinished.

One of the benefits of the platform is to be given “Prioritisation in people search results, so people can find you more easily” (as stated on the VIP sign up page). I don’t know how true this actually is as I can’t make comparisons to my status before the program.

I do know, that those following is a factor, type ‘Jonathan’ into periscope and i’m 10th, the results are in ranked order because of following. All twenty in the search result are twitter verified with myself in the middle. If I type ‘jonathanjk’ all my accounts show up regardless of their VIP status in the top twenty.

If I type ‘Hong Kong’, I don’t show up in the top twenty, the placement depends me putting ‘hong kong’ into my profile. Just like followers is being counted (and number of followers being counted as part of the search placement results is dynamic, if I over took Jonathan Ross from British TV, I’d be first), I think location should as well, without me putting it in my description. Changing my profile description to include ‘#hongkong’ actually immediately places me 6th on the list.

What also caught my eye was my New York Times editor friend, he’s ninth on the list and never scopes, he has over 1k following because of his Twitter verification and twitter presence (huge). A lot of Twitter verified users do show up in the rankings, while I have no beef against that on the surface. I wonder how active those accounts on the whole are. My friend’s isn’t at all and it got me thinking, if following numbers have an impact on search results, why can’t the regularity of when someone broadcasts have an impact as well? Surely if a person is interested in following a scoper from Hong Kong and wants to know more, does the platform benefit by throwing up accounts that are hardly used? Surely active accounts should get a higher ranking. How often a scoper broadcasts is even in the requirements for the VIP program.

Periscope is in the business of connecting viewers with broadcasters, so do it!

Now does any of what you’ve read help with the perception of quality broadcasts? No, this is just one piece of an already large pie when discussing discoverability. We will return to this in part 3.

The PET (Periscope Editorial Team)

It strikes me as odd that the Periscope team feature types of broadcast that have a limited appeal to viewers or regularly feature similar looking scopes. Looking at the watch time would illustrate the video featured isn’t of interest and I’m not sure how many sunset scopes or lovely walks in the country side I can take anymore, .

Simply put. I have five minutes, I’m looking at my home-screen, where do I use those minutes? On the YouTube app where an algorithm gives me what I want, or a live video (if it’s still live) of something a human has curated for me on Periscope. The live experience counts so much, but typing is extra work on Periscope, for YouTube, it’s an after thought.

It’s been made clear to me from a member of the team at Periscope that Periscope won’t feature replay broadcasts. I would argue against that somewhat as YouTube is mostly all replays. Second, are the 6-7 scopes featured, really the best of what the Periscope platform offers? With featured scopes also staying on my phone for roughly 24 hours. Do you know how fast YouTube works to cater to my flippant tastes on its home-screen?

YouTube is All About the Replays!

I want to suggest a doubling of the number of featured scopes. To save time for the viewer, split them across two feature boxes as opposed to one. Later, with time, feature a greater variety of broadcasts and dedicate the feature boxes to single categories? With watch times in these scopes amounting to 17 seconds to a 1 minute, there needs to be more featured content because people drive through them almost instantly. The average watch time proves this.

To further this suggestion of featuring more broadcasts, (in order to better flesh out the app, and keep people onboard) VIP scopers are somewhat at a loss with the fact that Periscope makes no effort to help scopers with providing residual value to their replays (how can they be shown again). Though to defend Periscope first, it wouldn’t hurt broadcasters to figure out a way where they the broadcaster themselves could indulge in a little bit of self-promotion by their own means. I know Periscope is about the live experience first, but this only goes so far towards entertaining users when there are only so many live scopes at any one time. Why can’t their be a‘Remember This?’ or ‘You Missed This’ feature which do in fact show replays. Is there really any harm in having a replay section to feature more entertaining scopes? To provide even more variety Periscope has to look back for great replays that were never featured live. If you want to get technical, featured live scopes are featured longer on the replay than they are live anyway.

Broadcasters can be happier because there is a potential for further discovery and second, there is more content to keep people on the app. The editorial team in theory should be able to pick even better scopes because the scopes don’t need to be live and you can use ‘watch time’ as a metric for scopers who want to submit broadcasts to this section of the app.

To Conclude

There isn't really a conclusion to this as part 3 directly continues and concludes on part 2. Part 3 will directly address the issue of quality broadcasts through a more in-depth discussion of watch time.

----------