Feed
iPhone et al
I haven't written much lately, so I thought I'd be controversial and write about what I'd like to see in the next iPhone and iOS7. I'll start with the easiest one, the iPhone:
Windows 8
I've been using Windows 8 quite a bit lately, so I thought I'd actually review it, rather than simply moan about it in 140 character chunks. So, without further ado, here we go:
Pros: Seriously, it's not all bad; there are some good bits. I like, for instance, the lack of Aero. The new design is much nicer, and it makes the OS smoother as well. The design elements that come with this UI change are nice as well, like the new progress bars, buttons, centred window titles, and default font. The startup's quicker, which helps to counteract the length of time spent working around the Start Screen. It's still not as quick as OS X, but it's significantly more responsive off the starting block than its predecessors.
I like the parts of Windows 7 that have made it through the mincer; the desktop mode includes the snapping features and peeking, and the in-icon progress bars are useful. I appreciate, both as a user and as a developer, the improvements in IE10, and don't particularly mourn the loss of the shiny logon page. It's also cheaper than 7 or Vista, but that hasn't really worked to increase sales spectacularly. Some of the applications, services, and support features launched alongside it, however, really are useful. I like SQL Server 2012, and find it somewhat easier to use than its predecessor. Similarly, Visual Studio 2012 is really nice, and lots of new features keep getting added in frequent updates (this does have its downsides). I also like the fact that the taskbar is replicated on dual screens. In nearly every instance, this is lovely. Some applications of dual screens won't want this, but it's easy enough to hide, I've found. Also, the progress dots that obey gravity are nice, even if the transition is a little botched.
Cons: Let's start at the more trivial, and then get into the bigger ones later, shall we? My first niggle may seem really trivial to the casual user, but it's actually fairly important to system integrity. Why do all users have to define a password hint? There is no reason that this should have to happen. I know my password, I know when I change it (regularly), I'll remember my new one. Why, then, should I have to write something that takes a huge element of guessing out of a brute force attack? Taking into account how regularly the same sort of passwords crop up, any casual attacker (say your laptop's been stolen by someone you know) can narrow that list down to three or four possibilities in one foul swoop. As such, I've worked my way around the requirement by typing "null" as my hint.
Similarly trivial, but an issue for which I have downloaded a applet to resolve, is the lack of certain features that OS X has. Reverse scrolling, for instance, makes sense in a system that is usable as a desktop and tablet OS, but no, it's not there. If you've got Windows 8 on one of those flipping sometimes-tablet-sometimes-laptop machines, then you are switching between directions when scrolling in different modes. Don't worry, there is an app for that, but only, I fear, if you've got a bit of know-how (or an internet guide) and access to the desktop mode. It's called AutoHotKey.
Next are my niggles about speed. To begin with, it takes an agonising few seconds to wake from sleep. I've got Windows 8 Pro, installed on the latest version of the MacBook Pro, running a Core i7 processor, 4GB RAM, and a respectable 5,400rpm HDD. OS X can start from sleep in the time it takes for me to open the lid (just over half a second). Windows, meanwhile, takes a good 2 seconds longer than that, and then I have to hit space before I can type my password. That lock screen thing is going to be the death of my space bar. I've also set my desktop background as that lock screen image, but the majority of my background is a creamy-white colour. As such, you can't really see the time display on it. It's not too much of a stretch of the imagination for the OS to recognise a lightly coloured background, and invert the time display to make it stand out. As I'm only on that screen for long enough to remember how annoying it is that I have to go through it every time I wake it up, I never put effort into discerning the greyish text from the creamy background. However, the first thing I'll do on unlocking is check the taskbar clock.
On the issue of speed, has anyone tried searching for an application? I'll not get into the fact they're basically undiscoverable in the Start Screen yet, but I immediately go for typing the first few letters of the name. That's how I do it on OS X: I hit the bottom of the screen to initiate Launchpad (strangely, I have always had the bottom left hotcorner assigned to this, even before Windows 8), and type "dr". That'll narrow my options down to Dropbox and Dreamweaver in an instant. However, search in Windows 8, where comparatively fewer applications are installed, and the search feels muddy. I think it's partially that there is that annoying animation there, but there's no completely instant feedback for your search. All I've got in Windows 8 is Visual Studio, the SQL server and its tools, Chrome, Blend, and 3 Bootcamp tools. I've uninstalled the package of apps that came with the OS, but I've also enabled the system tools' tiles. However, it's still a slower process finding an launching an app in Windows 8 than it finding one of the 126 applications installed on my iMac. That's partially due to the faster searching, but also the fact that magical, mystical tools in the form of folders exist.
Right, now onto system tools. Why is it not easy to enable a shutdown tile by default? Without building your own shortcut, you have to move your cursor to the bottom right, wait a bit, and then make three clicks. On older versions of Windows, it was one keyboard tap, and one click. Why do the network options in the charms bar not provide a simple way to open the advanced network settings like proxies? If you are an uninitiated user, you'd think they weren't available, or get stuck and put more stress on your poor technical support team. All for the sake of one button. Why does the inbuilt Mail application require a Microsoft ID before you can add a non-Microsoft mail account? Why, for that fact, did it break irreparably when I tried to set up my iCloud using server details that Outlook is perfectly accepting of?
Final niggle on this front: Why do developers have to ask your permission, as a desktop user, for their apps to access the internet? Could there not be a checkbox or other system identifier that decided that users who aren't paying through the nose for mobile internet will always allow their apps to connect to the internet. Or use the camera. Or touch anything outside their padded cell. I could, for instance, build a desktop application that opens an internet connection and passes data to a database on my server, syncing your details with other machines, sites, and accounts. I could download data stored in the cloud as and when it's needed, rather than placing gigabytes of soon-outdated definitions onto your machine during installation. You could then switch to an app version of my software, using the same computer, internet connection, and details, but you'd be prompted to allow internet access. You might not know that I've been downloading the odd kilobyte now and again, keeping your data relevant, and fleshing out your user experience. A good programmer can make this interaction with local and server-based data feel fluid and seamless. However, you might choose not to enable the internet connection rights, even though you were fine with them a second ago in desktop mode. Now, I have a choice as a developer. Do I include massive definitions in my app to combat people doing this, and then constantly roll out updates so you're always relatively up to date? The only other option would be to refuse any service to people wary of accepting that in-app connection. My pet hate is constant updating, where you cannot use an application because it needs an update, or your use is limited by the fact that you last connected to the internet 5 seconds ago, and something's changed in that time. Anyone using Windows 8 in both desktop and app mode should not have to specify privileges to apps that desktop software has by default. It confuses the user, makes more work for the developer, and potentially means users are downloading massive amounts of data for no reason.
Now to the main issue for discussion today: Start Screens and mistaken identity design choices. I get that it's meant to be a diverse OS. I get that people use it on tablets or touchscreen devices. I'm sure some of the choices made by Microsoft are lovely on such a platform. However, if your platform of choice is a desktop machine, as most Windows users are, then it's just time consuming. you can't just hit start, and be typing something into the search, or opening another application, at the same time as seeing what's currently on screen. You can't tap a couple of buttons and have a completely different tool available to you without disturbing everything else on screen at the same time. This is the main reason Windows 8 hasn't hit off in the workplace; it's not a workplace OS. It's trying too hard to satisfy too many consumers, and it's leaving out its main source of custom. The last thing a serious network manager wants to do is create unnecessary work for their team, and the employees who use their network. However, they now have to choose between up to date security and applications, or giving users something they know how to use. It's a shame really, because the pros I've given above make the basis for Windows 8 something that is so enticing. I design desktop applications in Windows 8, knowing my clients will be looking at the entire thing through the hazy gloss of Aero, instead of the clean, sophisticated Modern interface. Some of the tools available in the background of Windows 8 would improve the security, simplicity, and usability of systems from single PCs to whole networks, but it's held back by the fact that it's a daunting, new, and broken experience for users.
The best thing you can do, if you are a Windows 8 user, is install a tool to reinstate the Start Bar. It's not a perfect solution, but it helps. You still have niggles of the awkward UI left over; you can't hover over the close button in a maximised desktop application without those infernal charms appearing. Admittedly, this does not affect the usability of the close button, but it gets in the way of the window content. Also, if you're deploying on a network, you don't really want to apply a tweak like that, because then your user's will go home, install Windows 8, and moan that it's broken because the Start Bar's not there. Rather than having to deal with friction caused only by differing OSs between home and work, you have to deal with the different ways of using those OSs as well.
Conclusion: It's not polished, I can tell you that. There are some seriously great elements inside Windows 8 that are lost under the serious usability issues, workflow flaws, and insensitivity of Microsoft towards desktop users. With some tweaks, it can become much nicer, but they don't deal with the underlying broken concept on which Windows 8 is built. I will, sometime in the near future, collate a list of applications I think work nicely to make Windows 8 a more friendly OS, but I'm still trying a couple of options out on some fronts.
Sent from my Mac.
Tortoise
My tortoise has woken up. That's basically the central theme of this blog post. There's not much else, really. Anyway, it's still bitterly cold here, but it was time for him to wake up from hibernation, and we put him in an indoor hutch to see if he'd wake up. He did, and has been happily wandering about for a day. As I can't do anything without technology being involved, however, I've (with the help of my brother's apparently limitless inventory) rigged up a camera feed inside the hutch. You're more than welcome to watch him sitting around, doing very little, live on the internet. You may now get back to whatever you were doing. Also, Happy New Year or whatever else may be happening when you're reading this.
How much does it cost to send an email?
Let’s start with the basics: internet connection. Now, for the sake of this exercise, I’ve written an email consisting of 5 paragraphs of lipsum text. This works out at a nice length, and fairly representative of emails I send. With my customary signature, this email works out at 8558 bytes in size.
I tested my internet connection, and the upload speed came out at 7.32Mb/S. This is the important part for my example, as I’ll be sending the email over my connection, rather than receiving it. There are 8 bits in a byte, and 1024 bytes in a megabyte. With a little rearranging, one byte works out as 7.6294x10^(-6) megabits (Mb). With that in mind, my email will equal 0.065Mb.
Now, my connection can manage that data in under 1/100 of a second. However, the limit of my connection is not the limit of transfer. To test the rate at which my system sends this data, and to get a better idea of what happens, I monitored my connection during sending. My data sending rate peaked at 19KB/s, or 152Kb/s. However, it went from 0Kb/s up to this during transfer, so I’ll take an average of this, which is 76Kb/s. Now, at this speed, it’ll take about 0.88s to upload the email. This is the speed figure I shall work with from now on.
I have an unlimited broadband package that costs £7.50 a month. A month averages as 30.44 days, which translates as 730.56 hours, 43833.6 minutes, or 2630016 seconds. This puts my package at 0.0003p per second. However, my supplier stipulates that line rental must also be purchased from them at £14.50 a month. So, the monthly fee for my internet connection is in fact £22, or 0.0008p per second. To give you a bit of context, I’d need to send that same email 1250 times to spend a single penny.
However, that’s not where this ends. That’d be far too simple. Because an email costs more than that to send. Next, we need to take into account the cost of running the equipment needed to send the email. Again, I’ll be using my baseline equipment for this example. The first thing I need is a computer. Power consumption for these is hard to measure, because it depends what you’re doing. However, this site gives me a starting point. It’s not using the exact model MacBook Pro (MBP) I have; mine is newer. I don’t know how the newer one compares, but I’m going to put it in the area of 16W (watts) when typing and sending an email.
Now, I need something to power my internet connection. My ISP provided a router with their package, which I use as a modem. It’s based on a Netgear system, but specifics for the customised model that I own proved difficult to track down. However, I found this site, which gave me a ballpark figure of 10W for a similar router. I don’t use the access point function on mine, however, so I’ll knock a few watts off. 5.5W in fact. This value is taken from my my next source, and is the powered on consumption of a Netgear access point. That same source gives me a value of 4.1W that’s needed to run the AirPort Express base station that I use for WiFi.
All together, this gives me a total power consumption of 24.6W. However, I don’t technically use the modem or access point whilst I’m writing the email. So, I can write the email whilst using 16W of power, and then boot up my connection to send the email. I did a test to see how fast I type, and the result was 54wpm. I have to type 425 words of lipsum to make up the content of my email (including a short subject), so that will take me 7.9 minutes. I tested how fast my MBP could start, and came out with about 18 seconds. My source for MacBook power consumption suggests it uses 30W during startup, so I’ll take that into account. As such, my power requirements for sending an email are:
New Stylesheet
If you're reading this, then you've seen my new stylesheet. Not much of the actual content has changed, but I've made it simpler to navigate, and easier to read. A major part of this was simplifying the way content was laid out, so the About and Services pages now filter content down to only what you choose to see. The backend generation of the site has also been streamlined, so serving up content is faster and takes less of your computers' resources.
The site is also more navigable for those with different accessibility requirements, and content will be handled better by screen readers. I'm also going to be adding a few more accessibility options over the next few weeks, so you can adapt the site if you have issues reading with the colour scheme I have selected. I've also improved the way 404 errors are served to users of my site, and those of sites that I manage. All of these will be delivered to a friendly error page, with personalised options that help you find what they're looking for without interrupting your flow too much.
Some of the features on my site are now developed with Adobe Edge Animate, a wonderfully powerful application that I really like. It replaces nearly every instance that Flash would have in the past, and helps us step closer to a plugin-free web. As an Apple fan, my site hasn't ever used Flash (that I can remember, anyway), but it's still taken time for me to make sure it's as compatible with all devices as possible. This new stylesheet, and the javascript that goes with it, has added fallbacks for devices or browsers that don't support my original vision. Hopefully, it's seamless enough that it isn't too noticeable. The only page of my site that doesn't currently support the new stylesheet is the page that lets me write these posts, so I'm writing about a stylesheet I can't even see at the moment. Anyway, I hope you like it, and look out for a couple of minor changes that are still to come!
