In Singapore, getting your GST back is efficiently electronic

We departed Singapore and got to experience their electronic tourist refund scheme (eTRS), for collecting your GST back. To think, at the cashier’s desk when we made the purchase, we were wondering why we couldn’t just use a Global Blue refund card (useful in most of Europe). 

At the point of purchase, you are given a receipt with a barcode. When you visit the airport, just follow a touch-screen based interface that says you accept the conditions (i.e. this stuff is for export/you’re not Singaporean), swipe your passport (yes, it reads it all very well), say when you entered Singapore (you can get a popup calendar), scan the receipts you’ve received via the barcode (which will display the store you purchased things from, etc.), choose a refund method (we chose to just swipe a credit card), and voila! you get a notification receipt saying all is well. There was no physical inspection required, and with the other 5 people around, none of them had that requirement.

Apparently the monies get refunded back to the credit card within 10 days. This is extremely efficient – compared to even checking out in the UK or Europe. There you still have to get stamps on receipts, usually by lining up in a pretty long queue, then posting stuff back. 

The efficiency definitely leaves a very good aftertaste.

Do you need the .com?

Conventional wisdom says you need to have the .com in a domain name. Nowadays its backed by the fact that you have the .com easily available on your mobile phone’s popup keyboard as well.

But lately I’ve seen some quality sites launch and they don’t have the .com’s to go with it. AppleWorld.Today. Fusion.net. In the past, let’s not forger <re/code> from the AllThingsD folk at Wall Street Journal (also, .net). You’ve always had John Gruber’s Daring Fireball in the .net-namespace.

So maybe you don’t need the .com and you’ll do just fine.

It’s interesting to travel to places like Iceland where its common to see domains with .is. They’re happy using their country code top level domains (TLDs). It parts of Barcelona, its quite common to see .cat. 

Microsoft’s reaction to open

It’s interesting to follow what Microsoft has been doing, especially in relation to their reactions to the open world.

The One Laptop Per Child (OLPC) project was always interesting. It ships with Linux as a default, and it didn’t take long for Microsoft to offer Windows on this device. Though its likely that there were never any large deployments for this.

Now, you find that Microsoft Windows 10 is basically a free upgrade for users from the days of Windows 7. And it looks like upgrades are going to be free going forward. This follows what Apple has done with OSX, and if you look back even a decade, this seemed like an odd choice – there was money to be made with the OS. This will adjust some TCO calculations for sure.

Now there’s the Raspberry Pi 2. It is also the first Raspberry Pi to be able to run Windows for IoT devices. Apparently this will also be free (for makers; whatever that means). You’re buying a $35 computer, and using the SD card as an easily swappable OS.

I should probably also mention that you can run Linux on Azure, their cloud platform.

So all in, it’s interesting to watch Microsoft, the company once the epitome of being proprietary, now learning to embrace open.

Here, and defending your trademarks

I read: Nokia threatens London start-up over ‘HERE”.

It’s all about Lowdownapp (I’ve not heard about it before this), made by David J Senior & crew. I think the crux of the problem is that they have also released an app called HERE and Nokia is obviously pissed because of HERE Maps.

Apparently Nokia has spent USD$12m on creating the HERE brand and are now defending it.

I’m not surprised this startup hadn’t heard of it. I’m also not surprised that unless you’ve used a Windows phone, you’ve probably not heard about HERE Maps either. There was a release of HERE Maps on iOS, but I’m sure it never got the attention that Apple Maps or Google Maps got (I’m including Apple here because laugh as much as you want, being a default, really helps).

A few months back, I spoke to an entrepreneur doing indoor mapping, and mentioned to him that Google Maps is starting to encroach on that space as possible competition (he knew that). I then said that the best indoor maps I’d seen so far had come from HERE Maps. He had never heard about it, and he’s deep into mobile and mapping. 

It’s a sad fact of life that $12m is money not well spent, because no one at the moment really cares about the Windows phone platform; so if that is your app showcase, you’ve screwed the pooch. To make matters worse, you can’t even find HERE Maps on the Apple App Store today (it was pulled down in 2013, but apparently will make a comeback in early 2015). It is still available as a beta in the Android Play Store. I liked this thread between Benedict Evans and David Senior, because while its clear that Nokia does have a trademark, and its clear to Benedict Evans since he watches this market, its definitely unclear to the masses that Nokia has anything to do with Here.

Should all startups perform a trademark search before naming their companies? I’m not sure – its already hard to get a good name with domain/social media presence these days. Plus its time consuming (not to mention costly) to do a trademark search in multiple jurisdictions that you care about (or maybe you can use a service like this?). Not something the average startup wants to spend costs on.

Heck, even established companies like Microsoft end up doing a rename of SkyDrive to OneDrive to please BSkyB. So it’s not a startup rookie mistake either.

What happens next is likely that the folks at Lowdownapp will rename their Here app; the functionality I’m guessing will remain the same, it will just be called something different. Do I like it? Absolutely not. However as a bonus, it looks like the app only launched in Dec 2014 so maybe it will be easier giving in to this battle.

MariaDB turns 5!

I stopped working on MySQL at Sun Microsystems in late 2009 (after a lengthy period of garden leave), to join Monty Program Ab, and was greatly anticipating a MariaDB release that we could take to market. The first GA release of MariaDB came out February 1 2010 – MariaDB 5.1.42. Today is MariaDB Server’s 5th birthday!

We didn’t even want to call it GA back then – we referred to it as a “stable” release. We didn’t make our own builds because we figured source code tarballs were good enough; so builds were made and hosted at OurDelta. It took some months (around August 2010) when we moved release notes to the Knowledgebase (which you’ll notice has moved from kb.askmonty.org to its current location) from the old front page wiki install that we had at askmonty.org.

I didn’t go to the first company meeting in Malaga due to having the chickenpox, so my first meeting was the one we did in Reykjavik, Iceland. We did it towards the end of February 2010, and planned it literally in a month – maybe a celebration that we brought 5.1 to market on time, and also to plan 5.2.

Speaking of companies, we were Monty Program Ab (professionally this quickly became MariaDB Services Ab), then SkySQL Ab (via merger), and finally MariaDB Corporation Ab (via re-branding). Shortly before the SkySQL Ab merger, we even have the MariaDB Foundation appear.

Anyway, what have we released? MariaDB 5.1, MariaDB 5.2, MariaDB 5.3, MariaDB 5.5, MariaDB 10.0, MariaDB Galera Cluster 5.5 & 10.0, a special MariaDB 5.5 with TokuDB build and a special MariaDB with FusionIO improvements build. To boot, we also have three client libraries (connectors, if you must): C, Java, and ODBC.

So 5 major server releases (7 if you count the Galera series), and we’re now working on MariaDB 10.1. I count 88 releases of the server across various versions (with breakdowns: 9 alphas, 11 betas, 7 release candidates and 61 GAs). We’ve had 23 Galera releases and 15 releases for the various client libraries.

We are shipping in all major Linux and BSD distributions. In many, we are even the default

This birthday is a nice time to look back at our achievements, but also to remind ourselves to not rest on our laurels and continue to focus on growth. The last sanctioned press release talks of over 2 million users globally. 

Thank you to all our users. Thank you to all the contributors and developers. Here’s to a lot more adoption, growth, releases and technology improvements!

Samsung should learn to take care of their customers

My first Samsung mobile phone was a Samsung Galaxy S3. I still use it as my roaming phone, though mainly all it does now is serve as a device that I plug into a battery pack and let it act as a tethered modem.

The main reason is that its slow. Its old. The software on it is outdated. Keep in mind this was a phone released in May 2012 (I must have gotten it in June 2012). The equivalent iPhone that came out in 2012? The iPhone 5. The iPhone 5 still runs iOS 8 without issue. Samsung is notoriously delayed when it comes to updating software; in fact they rather you buy a new device. No, the average user isn’t going to go around hackery to get the latest Android on it.

So when I see articles like Samsung profits down as smartphone division feels squeeze, I can only chuckle. Knocking off iPhone was a good model, but then you see Xiaomi come into play and do the same thing at 1/3 or 1/4th the price. Updating software – Xiaomi does it every Friday. Samsung requires you to heavily pray for an update, or pay to get a new phone. 

High-end smartphones in Malaysia cost an average of RM2,000. You can buy a computer at this price. Computers typically have a 3-year warranty, and get software updates for 3-5 years. Mobile phones come with a meagre 1-year warranty, if you buy it on plan, it is a 2-year lock-in, and by the time your lock-in is over, you’re buying the next phone (very unlike a computer, eh?). This is why its smart that Apple does iOS updates for years on end (I reckon they focus on 4 generations at any given time).

It’s also interesting to watch the secondhand market. See what last year’s model of an iPhone sells versus a Samsung. 

Samsung needs innovation. It needs leadership. It needs to learn to be more open. 

Will I buy another Samsung phone based on my S3 experience? No. Have I seen any Samsung Note users migrate to the Apple iPhone 6+ yet? No. But again, it is still early days in 2015.


i