Google – Flavour of China.

RFM AvatarSmall

 

 

 

 

 

Getting back into China will be long, tough haul

  • Google has been effectively absent from the Chinese market since 2010.
  • With the growth of Internet use in China and ever increasing smartphone penetration, it is clear that this is a big revenue stream that is currently passing it by.
  • Looking at the ecosystem opportunity, RFM estimates that the Chinese market could account for 30% of all users by the end of 2016E.
  • With developed market growth slowing, this is an opportunity that Google cannot afford to ignore and this is why I suspect that it is looking to enter this market once again.
  • This time it looks like Google is going to make its Google Play or a version of it available in the Chinese market.
  • This also co-indices with a move to allow developers based in China to make money in 130 countries but with the unfortunate exception of China.
  • How this would work is very unclear but I think it very unlikely that this is the beginning of free availability of Google Play on all Android devices.
  • Google Play is Google’s most precious mobile asset as it is the availability of the apps that users most demand.
  • In most markets, in order to get Google Play a handset maker or and operator needs to sign an agreement with Google.
  • This ensures that all of Google’s Digital Life services are also present on the device and that certain criteria around placement and defaults are met.
  • This is how Google’s ensures that its ecosystem is present on around 40% of all Android devices in the market.
  • RFM estimates that Google’s ecosystem will generate around $6.9bn in mobile advertising revenues on Android devices in 2014E.
  • Consequently, I suspect that a Chinese store will be something very different to Google Play, probably with a totally different name.
  • I also suspect that it will be very Chinese in its nature and mostly feature apps and content that are tailored to the Chinese market.
  • Hence, this is not going to represent a backdoor way to getting Google Play on a device without deploying the rest of the Google Ecosystem.
  • The Chinese ecosystems are at an early stage of development and I am looking for three big ones to emerge owned by Baidu, Tencent, Alibaba and maybe Xiaomi.
  • I think that these players are intent on keeping Google out of their home market and are likely to keep their popular services and apps out of whatever Google launches in China.
  • Consequently, I see no change to the existing status quo and continue to expect Google to see no real increase in traction in China.
  • Despite that, the revenue outlook picture for Google still looks healthy and the stock is more than fairly priced despite the overspending and the shortcomings in corporate governance. 

Yahoo! and Mozilla – Face not wallet

RFM AvatarSmall

 

 

 

 

 

The loss of Firefox hurts image and more than revenues.

  • Yahoo! and Mozilla have announced that Yahoo! will replace Google as the default search engine within the Firefox Internet browser.
  • This change marks the end of the 10 year deal in which Google was the default search provider for Mozilla.
  • Mozilla and Yahoo! have signed a 5 year deal where no details were disclosed.
  • Through the Firefox browser Mozilla was acquiring traffic on Google’s behalf and its share of those revenues made up the vast majority of Mozilla’s revenues.
  • In 2012A (last available figures) Mozilla had $311 in revenues where I think Google was probably responsible for around 85%.
  • Google typically pays away 20% of its gross search revenues to those that source traffic on its behalf implying that something in the region of $1.3bn of its gross search revenues was coming from traffic sourced from Firefox.
  • In 2012A Google’s gross search revenues were $37.6bn meaning that being the default search provider to Firefox represented only 3.5% of revenues.
  • Consequently, I suspect that this means that the loss of Mozilla is unlikely to have much impact on Google’s ability to earn revenues from search.
  • This is especially the case as since 2012A, Firefox’s desktop browser share has declined from 22.5% to 14.2% while Chrome’s share has risen from 17.6% to 21.1% (see here).
  • Add this to the fact that Yahoo! is desperately trying to gain share in search, and it is not difficult to see how Yahoo! was willing to pay more to be the default search provider than Google was.
  • Furthermore, I suspect that Mozilla was keen to get away from Google given that it is actively trying to lure users away from Firefox to its own Chrome browser.  
  • As a result Yahoo! clearly makes a better match for Firefox than Google does and both parties had an incentive to see a deal done.
  • Hence, Yahoo! is probably paying much less than some commentators may think, but probably more than the 20% gross search revenue share that Google was paying.
  • Whether, this transaction meaningfully lifts Yahoo!’s search revenues has yet to be seen as one must not forget that Google is better at monetising traffic that comes in than Yahoo! is.
  • Net net this is a good deal for Yahoo! and it should have some top line impact but the loss to Google is probably more face than wallet.
  • I would still pick Google over Yahoo! in the short term, given Yahoo!’s inability to execute on the assets it has already acquired.

Yahoo! and Aol – Heaven and Hell

RFM AvatarSmall

 

 

 

 

 

A deal with Aol would be great for the stock but bad for the company.

  • The Aol and Yahoo! story is heating up again (see here) as Starboard has gone back into Aol and now has a reasonable stake in both companies.
  • There is substantial value to be released in both companies and while putting them together would give a one-time release of value, it would do nothing for the long term outlook for either company.
  • The main issues here are that Yahoo! is trading at a substantial discount to the sum of its assets and that its core business continues to underperform.
  • The activists want Yahoo! to realise the value in its Asian assets, cut costs and return cash to shareholders thereby releasing substantial value for shareholders.
  • Such a release of value would most likely result in a one-time substantial rally in the stock but little else.
  • They are also pushing Aol and Yahoo! to get together in the belief that they can solve each other’s problems.
  • Unfortunately this approach ignores the fact that Yahoo!’s only real chance of seeing real growth again lies in the development of its ecosystem.
  • Over the last 2 years it has accumulated a series of Digital Life assets that together give it over 70% coverage of the Digital Life Pie.
  • While it has been great at accumulating these assets it has done almost nothing with them.
  • These assets need to be integrated and rolled out on mobile in a fun and easy to use way where users can identify with Yahoo!
  • That way users would want to spend time with Yahoo! which Yahoo! can then use to sell valuable, targeted advertising.
  • Unfortunately, none of this has happened which is the main reason why the core business has been a bad underperformer this year.
  • Getting together with Aol does nothing to fix this problem and in every likelihood would make it worse.
  • This is due to the added complexity of adding in Aol’s assets on top of the jumble that is already there at Yahoo!.
  • Furthermore, other than the potential to release value for shareholders, this combination does nothing for Yahoo!’s ecosystem strategy and nothing to get growth going again.
  • This is why I suspect that the management of Yahoo! will resist this deal but the sale of its Asian assets would go a long way to mollifying investors.
  • I continue to be frustrated by Yahoo!’s lack of progress and think that until there an improvement in the company’s ability to execute, its core business will continue to underperform.
  • I am taking Alibaba-related profits on Yahoo! as there is no sign of a turnaround and I do not expect it to sell its Asian assets.
  • I could be looking for a re-entry point in 2015E but I need to see some signs of progress first.  

 

BlackBerry – Commodities trader

RFM AvatarSmall

 

 

 

 

 

BlackBerry now looks stable but commoditising fast

  • Blackberry held an event in San Francisco where it unveiled the new version of its server product (BES12) and launched a partnership with Samsung for regulated industries.
  • The new version of the server makes a number of meaningful improvements over BES10 that include:
    • First: Seamless support for all platforms.
    • The platform supports iOS, Android, Windows Phone, BlackBerry 10 and BlackBerry OS.
    • The ability to run both versions of Blackberry in the same place is a meaningful improvement as until now the two different versions had to be managed separately.
    • Second: The platform supports all ownership structures including: BYOD (bring your own device), COBO (company owned business only) and COPE (company owned personal enabled).
    • Third: the platform has designed to be saleable with up to 25,000 devices per server and 150,000 devices per domain.
    • Fourth: BlackBerry has placed its development emphasis on user experience, extensibility into Internet of Things (IoT) and total cost of ownership.
    • Fifth: A number of advanced features such as VPN authentication, simple and secure authentication among other features were also launched but only for BlackBerry devices.
  • At the same time, Samsung and BlackBerry have jointly announced that BES12 will support Samsung Knox and that the two companies will work together to address the regulated and government sectors.
  • These are two sectors where security is incredibly important and where Android’s lack of security has hampered its adoption.
  • Samsung Knox is also thought to be not that secure and teaming up with BlackBerry will help Samsung penetrate this industry more deeply.
  • I suspect that this relationship will be along the lines of offering a complete device and service offering for these industries.
  • Here, I think that BlackBerry devices will address the COBO and COPE segments of the user base and Samsung devices the BYOD piece.
  • John Chen is doing the right things to keep BlackBerry solvent but what is emerging is a vendor of MDM services.
  • This is a tough and competitive landscape with Airwatch, MobileIron, Good Technology, Microsoft and a host of others all vying to manage an enterprise’s mobile devices.
  • BlackBerry still has an advantage here as it has around 50m subscribers which is far more than any of its competitors and its solution is still the most comprehensive.
  • However, BlackBerry is having to offer heavy concessions (EZ Pass) to keep customers happy and I fear that this segment is rapidly becoming a commodity.
  • Consequently, I suspect that BlackBerry should now stabilise but the days of high margins, high growth and a high multiple are long gone leaving the shares with a difficult and uncertain future.
  • I would prefer Microsoft as an investment in the Digital Work ecosystem.

Twitter – Hopes and dreams

RFM AvatarSmall

 

 

 

 

 

Far more changes are required to meet the long term goals.

  • Twitter made some bold promises at its analyst day giving investors some confidence and sending the shares up 7%.
  • Firstly, a series of product announcements aimed at increasing engagement for logged-on users and casual visitors that include:
    • Instant timeline. Gives new users and casual visitors an idea of what is going on without having to follow anybody right away.
    • “What you missed”. Curates a list of tweets that were posted since the user logged in.
    • Homepage. A new design to make the homepage more engaging for both logged in and casual visitors.
    • New apps. New apps for mobile beyond Twitter and Vine will be released although what function they will serve is unclear.
    • Localised and event content. Content will be location aware and also curated by events that occurring around the user.
  • Secondly: Twitter provided some numbers that gave some context around the size of its audience and its long term targets:
    • Audience: Twitter estimates that its total audience is roughly twice the size of its user base at around 500m.
    • Long term audience: Twitter thinks that by 2020 it will have a reach of 2bn people, 4x where it is today.
    • Long term revenues: Twitter thinks that it can reach around $14bn in revenues within around 10 years, up from about $1.4bn in 2014E.
  • Thirdly: The company tried to play down the recent executive turmoil and turnover but in reality it explained very little.
  • Twitter is rapidly approaching saturation with its existing business and there has obviously been large differences in opinion in terms of where the company should go from here.
  • The fact that the head of product role has changed numerous times recently is a bad sign which only adds to the issues created by the departure of the head of analytics and VP of engineering at the end of October.
  • The medium term path has now been laid out and I hope that Costolo now has his ship under control and that turnover will now cease.
  • It will be a very bad sign if departures continue.
  • From the long-term goals it is clear that Twitter has no chance of meeting them unless there are fundamental changes.
  • I estimate that in its current form Twitter revenues will normalise at around $2bn.
  • This is because users only spend a very small portion of their Digital Lives engaging in microblogging.
  • Consequently, the information that Twitter receives about its users is more limited and the opportunity it has to advertise to them is also much shorter than for other activities like social networking.
  • Furthermore, Twitter knows nothing about at least half of the users that are viewing the tweets making targeting more difficult.
  • Some targeting will be possible based on the nature of the tweets being viewed, but the quality of these adverts and the revenues earned will be much lower.
  • In order to get to $14bn in revenues Twitter needs to address far more of the Digital Life pie and it must do so effectively.
  • Assuming it is a good instant messaging platform, Twitter currently addresses 9% of the Digital Life pie (see here).
  • If it can increase its reach to 2bn viewers on a monthly basis, I would estimate that Twitter needs to cover at least one third of the Digital Life pie to reach $14bn in revenues.
  • This means it will have to have a stab at competing in social networking, gaming or media consumption.
  • Of this, there is no sign and until there is I think it very unlikely that Twitter will ever come close to this target.
  • If it increases its reach to 2bn viewers with no changes to its current activity, revenues might just make $4bn in 2020E.
  • Either way it is clear that Twitter is not the next Facebook (see here) and while investors continue to pin their hopes on this dream, they are in for disappointments.
  • Twitter is a great company but I think the stock will remain bad for now.

Amazon Echo – No Eureka.

RFM AvatarSmall

 

 

 

 

 

Amazon Echo does nothing to address the big issues.

  • As long as Echo is a great Bluetooth speaker it is unlikely to be nearly as big a disaster as the Fire phone.
  • Amazon has initiated its latest experiment with the consumer with the launch of the Amazon Echo which is basically a Bluetooth speaker with a brain.
  • This is a large (9.3” x 3.3”) cylindrical Bluetooth enabled speaker that has a digital personal assistant built in.
  • The voice controlled device can be used to answer questions, set alarms, check the news and play music.
  • The device is priced at $199 which puts it in line with other Bluetooth speakers such as the UE Boom, Jawbone Jambox and Bose Soundlink Mini.
  • However, all three of these companies have invested large amounts of money to work out how to make small units make a large amount of high quality sound.
  • The Echo is far larger than any of them meaning that it will have to deliver better sound at a higher volume to get reviewers excited.
  • Furthermore, the device has no battery as it is designed for in-home use which is likely to count against it when the reviews come in.
  • The idea here is to give consumers easy access to various aspects of their digital lives (media consumption, search and reference) via Amazon web services.
  • Like Google, the Echo does all of its thinking in the cloud making it almost infinitely upgradeable in terms of functionality over time.
  • However this is where the problems will start to occur.
  • The personal assistant functionality of Echo will only be as good as the systems that sit behind it and here Google is far better than anyone else’s.
  • Recent tests of Google, Siri and Cortana showed that Google was far better at interpreting the questions and coming up with useful answers than either Siri or Cortana.
  • I suspect that Echo will lag both Siri and Cortana as this kind of artificial intelligence takes a lot of money and a long time to develop.
  • Consequently, I think that the Amazon Echo will live or die by how well it fares when compared to other Bluetooth speakers in its price range.
  • The fact that it has no battery will count against it but if it has top notch quality sound, it might just see some volume.
  • Finally, it is totally unclear how this product links in with the rest of Amazon’s ecosystem and how it will make spending time with Amazon more fun and engaging.
  • I continue to think that Amazon continues to put the cart before the horse when it comes to its ecosystem strategy.
  • The ecosystem has to be fun and easy to use before one starts making hardware to run on it.
  • Furthermore, the hardware pieces must all work together seamlessly and add depth and functionality to the Digital Life offering.
  • Amazon’s hardware does none of this.
  • It looks like a random series of experiments where no real thought has gone into what the products are actually trying to achieve.
  • Until Amazon gets behind a cohesive and all-encompassing ecosystem strategy, hardware will continue to hamper an already incredibly fragile bottom line. 

Microsoft – Digital work land grab

RFM AvatarSmall

 

 

 

 

 

Microsoft moves to cement Office’s dominance.

  • Microsoft is moving to tighten its grip on the Digital Work ecosystem before anyone else can really get a decent foothold.
  • Digital Work is the professional equivalent of Digital Life and describes the digital activities undertaken by users in the workplace and is measured by the amount of time each user spends engaged in activity.
  • One of the pillars of the Microsoft ecosystem is the marriage of Digital Life and Digital Work in a seamless way.
  • Given its dominance of Digital Work, this is a proposition that only Microsoft can make and in my opinion is a key differentiator.
  • However, Microsoft must remain dominant in this space and this is what all of the recent moves have been about.
  • These moves include:
    • Microsoft has now made it possible to create and edit Office documents on iOS and Android even without an Office 365 subscription. Some advanced features are only available for paid Office 365 users but these are unlikely to be required by the majority looking to do basic editing.
    • Updating its Office offering such that Word, Excel and PowerPoint work much better on iPhone, iPad and Android and committing to further regular updates.
    • Offering Office 365 subscribers unlimited OneDrive cloud storage.
    • Doing a deal with Dropbox to tightly integrate its online storage service into Office.
  • I think that there is one motivation behind all of these moves.
  • I believe that Microsoft wants to make Office so accessible, easy to use and useful that no one can be bothered to use anything else.
  • I also think that at the end of the day, the vast majority of creation and editing of office documents is going to remain on a device equipped with a keyboard and a mouse.
  • Touch based editing is fine for small refinements and proof reading but it is very clumsy and painful for detailed editing and for long typing sessions.
  • Consequently, I think it unlikely that Microsoft is going to lose meaningful Office revenues due to these changes.
  • On the contrary, it might actually gain as users are introduced for free and when they need to do more in depth work, they are more likely to become paying users.
  • This also creates a disincentive to use the inferior Google Docs or Apple iWork.
  • Microsoft is breaking with traditions of its past an adopting the models that are currently working with success in the Internet which is an extremely good sign.
  • This reinforces my hopes that Nadella’s vision of Microsoft as an ecosystem company if filtering down to the rank and file and that execution will be quick.
  • There is still a lot to do but all the signs are there that something new is emerging in Redmond.
  • Microsoft remains one of favourite places to invest (alongside Google and Apple) in the digital ecosystem.

Xiaomi – Reality check

RFM AvatarSmall

 

 

 

 

 

There is very little left for investors at $45bn. 

  • If Xiaomi can raise money at a valuation of $40bn-$50bn, this will be its greatest achievement to date.
  • Now that Xiaomi is the leader in the Chinese smartphone market, it is beginning to turn its eye to other potentially huge markets such as India and Russia.
  • Its strength at home lies in two places.
    • First: It sells nicely specified devices at excellent prices.
    • Second: Its media offering is very attractive to Chinese consumers and Xiaomi devices rack up more usage than iPhone in China does.
  • Offering nicely specified phones at great prices is a good way to get started but it is a disaster when it comes to making money.
  • It is through its second strength which represents the acorn of its ecosystem, that it has a chance to make some real money.
  • If it can make its ecosystem popular and demanded, then it can start to raise its prices and it is from there that the profits will come.
  • At the moment, I very much doubt that Xiaomi is making any money at all. If it was, it would not be raising funds.
  • I suspect that it wants to the money to invest in expanding into these overseas markets.
  • Its ecosystem as it stands today is very Chinese centric and to have a chance overseas it will have to develop something that is more to applicable to those markets.
  • If it has no ecosystem offering for those markets it will just be a commodity box shifter like every other Android maker.
  • That is going to need investment and this is what I suspect the money will be used for.
  • However, at a valuation of $45bn investors are already assuming huge success and upside from there looks challenging.
  • RFM forecasts that Xiaomi will ship 65m units in 2014E and 85m units in 2015E.
  • Assuming that the ASPs are around $250 (its flagship MI4 currently sells for $327), 2014E and 2015E revenues will be $16.3bn and $21.3bn respectively.
  • Xiaomi is considered by many to be very like Apple which is currently trading at 2.9x 2014E and 2.6x 2015E EV/Sales.
  • Applying this multiple to Xiaomi’s revenues one comes up with a valuation of $51.3bn which is how I suspect the current valuation of Xiaomi has been derived.
  • The flaw in this calculation is that Apple trades at this EV/Sales multiple because it has huge margins but Xiaomi does not.
  • Looking at its Android brethren, it is unlikely to make more than 4% EBIT margins unless its ecosystem is wildly successful.
  • In the best case scenario, this is going to take a very long time and there is a high risk that it never happens.
  • Assuming that Xiaomi never makes it past 4% margins then 2014E and 2015E EBIT would be $650m and $850m respectively.
  • Putting this against Apples 10.3x 2014E and 9.1x 2015E EV/EBIT gives a valuation of just $7.2bn for Xiaomi.
  • Xiaomi is of course growing much faster than Apple but even if I give the company a 300% premium to Apple to account for this, I still only get a $21.6bn valuation.
  • This is less than half the valuation that is being thrown around.
  • Consequently, money that goes in at $45bn is already paying for huge success in the ecosystem, monetised through high handset margins, all of which is extremely uncertain.
  • Hence, the main winners from a fund raising at $45bn will be the existing investors who will be giving away much less equity in return for the cash the company needs.
  • I would steer clear of this at any valuation above $21bn.

Sony Q2 14A – Tough times

RFM AvatarSmall

 

 

 

 

 

The journey to ecosystem remains very hard.

  • Sony reported a difficult set of results but there were no real surprises.
  • Q2 14A Revenues / operating profits were slightly better than forecast at JPY 1,902bn / LOSS JPY 85.6bn compared to consensus at JPY 1,812bn / LOSS JPY 62.7bn.
  • The main reason for the better than expected performance was the strength of the PS4 which has continues to substantially outsell XBox one.
  • Mobile devices marginally increased its share shipping 9.9m units which is equivalent to smartphone market share of 3.1% compared to 3.0% last quarter.
  • However the division recorded operating losses of JPY172bn as a goodwill write down of JPY176bn was taken during the quarter.
  • The write down was previously announced (see here) and excluding that the business made margins of just 1.3%.
  • The write-down reflects a much more cautious view on the outlook for the handset business and I suspect a reduction of long term margins from 10% to 3-4%.
  • This more cautious outlook is achievable as a shipper of commoditised devices, but I still think that more is achievable.
  • This is because Sony has the assets with which to create a proper ecosystem that appeals to users.
  • If it can improve the user experience on the PlayStation and migrate that across to its other consumer electronics products it will have a far more compelling proposition.
  • Furthermore, it also has media assets which it can put together with good Digital Life services to make users want to spend their time with Sony.
  • If this can be achieved then users will want to buy Sony devices meaning that Sony can once again charge a premium for its products and make some proper money.
  • Unfortunately, this remains very far from the today’s reality and a huge mountain is still to be climbed to reach this lofty ambition.
  • Furthermore it is going to be a painful road as it is clear that the investor base remains blissfully ignorant of what the company is trying to achieve.
  • Calls for the shutdown of the handset business and comparisons to Panasonic (which is now focusing solely on the enterprise) are good examples of how investors continue to misunderstand Sony.
  • With the exception of the Pictures segment, the other division of Sony have seen some progress with three of them returning to the black.
  • The outlook remains unchanged and I still see potential for Sony but a future driven by software and services requires huge change.
  • History shows that change can be especially difficult for Japanese companies I suspect that this will take longer than I am hoping. 

Microsoft – Band of brothers

RFM AvatarSmall

 

 

 

 

 

Microsoft’s Band is another shot at the ecosystem.

  • The new Microsoft Band is not much to look at but if it is robust and works as well as promised, then it could see some traction.
  • Microsoft Band is the latest is a slew of products that promise to monitor the key aspects of user’s daily lives and help keep them on the straight and narrow.
  • There are two main problems with almost all the pedometers, activity trackers and heart rate monitors that have been launched date:
    • First. They are not robust enough. Nike’s Fuelband, Fitbits and many others suffer from a relatively short life-span.
    • The review columns have plenty of complaints regarding the short life span of these devices and this true for almost all of them.
    • Activity trackers have to be able to withstand the hurly burly of an active lifestyle and if they break after a couple of months users are just going to give up.
    • Second: The sensors need to work on all people all of the time and be accurate.
    • The performance of heart rate sensors to date has been very far adrift of what is needed to be commercially viable.
    • This is why the Samsung Gear Fit does not offer continuous heart rate monitoring. It can’t hold the signal.
    • Accuracy is also critical. This is especially the case if the data is going to be used by health providers for monitoring, diagnosis and prevention.
  • If the Microsoft Band can address these two issues then it has a god chance of getting some traction.
  • Microsoft’s hardware products to date are surprisingly robust.
  • The Lumia devices, foldable mice and keyboards, while not water or coffee proof, are well made and withstand a lot of abuse.
  • The Microsoft Band makes no attempt at being waterproof, but its pedigree gives me confidence that it will be robust enough for the use case for which it has been designed.
  • The key will be how well it can pick-up and hold the heart rate signal.
  • The Band needs to be able to do this flawlessly on everybody all the time.
  • If it fails, then it will be useless as a fitness monitor, it will get terrible reviews and no one will buy it.
  • I am hopeful that Microsoft has made sure that this will work and assuming it does, then it has also a pretty good back-end lined up.
  • The Microsoft Band goes hand in hand with Microsoft Health which is very similar to Apple HealthKit and Google Fit in that it will collate, store and analyse all of the data regarding a user’s activities.
  • At $199 the device is fairly priced as long as it functions as promised.
  • Microsoft Band will work with all three smartphone platforms and launches with a range of partners and apps aimed at fitness.
  • This confirms that this is about developing the Microsoft ecosystem as a place to live one’s Digital Life rather than an enticement to buy software.
  • This is long game.
  • If users start spending more time with Microsoft services on Android and iOS then one day they might just be enticed to make the switch to a Microsoft product.
  • At the moment this is the only way that Microsoft will really be able to monetise these users as it has yet to go down the Google route of selling user data to advertisers.
  • The Band is not the sexiest product out there but if it works as advertised, it will be one of the best performers.
  • I remain positive on Microsoft’s ambitions in the ecosystem as it has the right strategy and buy in from the top to make it work.
  • The biggest problem remains the marketing message which still needs to be updated to address the realities of the market where it now finds itself.