Why it is important to get your website working well on mobile devices

Being a small business owner as well as a website builder (mostly technical) and owner of an entry level Smartphone and tablet PC has taught me important things about building a company website:

• Make sure that the core functionality of your website works without having JavaScript enabled. The reasons for this are as follows:
1. A lot of people where I live own basic Smartphones which often have JavaScript disabled by default. BlackBerry phones in particular are like this. Most of these people do not know about JavaScript or how to enable it.
2. Limited memory capacity: I had a case when I used to own a BlackBerry phone that a website would not open because it was super heavy and the phone ran out of memory. The site had lots of scripts and other busy stuff.
3. Device performance when scripts take up too much processing time. This gets so bad that the device responds slowly.
4. Let the JavaScript libraries load late, that way the page starts rendering early. Scripts can usually wait a second or two. Some scripts also take a long time to initialise while the user already tries to do something that relies on them to be ready. The best is to not having to require them in the first place.
   
• Allow the interface to change and re-arrange based on screen size. This also means hiding huge banners if a small screen size is detected. The min-width and max-width CSS options help.
  Example for a wrapper inside of CSS:
  @media only screen and (min-width: 650px) { }
  
• This is very important and is for all devices: Compress, minimize, cache, etc.
1. Let the browser cache images for months since they normally do not get changed once they have been created. The same applies to JavaScript libraries, CSS (once finalised) and other files, but be careful if your content changes, then that should not get cached.
2. Minimized CSS, JavaScript and HTML is always good to keep downloading times low. The browser also has less parsing to do if there are no extra spaces. Also consider that people with entry level mobile phones are also not able to afford bandwidth and you do not want to get rid of customers due to slow download on a simple device.
3. Get rid if any extra image data like comments, camera details and EXIF data. This information is totally useless unless you have specific image gallery sections that use this data. For interface images, you sometimes get lucky and strip off between 15% to 30% off the file size. Smaller files download faster and are likely to stay in memory longer due to more free space for extra files before being age flushed.
   
• Make sure the fonts are big enough for mobile devices and the interface looks good on a low screen resolution. Many websites look good when designed on a high resolution, but when switching to a lower resolution, then the ugly parts show up. The other way around is much easier. This also means your button text should not be too long or too short. These interface limitations are very likely to cause clutter and confusion, so you have to possibly break your site into smaller components and check each one individually. These components should be able to flow below each other or do something else on smaller screen resolutions without the user having to zoom in five times, preferably no zooming at all because zooming makes people give up quickly.
  
• Keep it simple: For search engines your pages should be specific, the same applies to the average user browsing your website. Too many call to actions (CTA's) will just be distracting and are overkill for low resolution interfaces. Enquiry and registration forms should remain simple since anything that looks remotely complicated will let users have second thoughts.
  I can recommend a book about web usability that I have read in regards to this: Don't Make Me Think

Another bad behaving IP address to block

Yet again there is some bad bot behaviour on my website. This time it comes from the IP address: 14.18.25.69
The behaviour seems to be a bot due to trying to use the external URL from Google Tag Manager as a URL on my website. The 360Spider already got a 418 HTTP response code today due to its nonsense.

Based on my quick research, the traffic comes from China. See:
http://myip.ms/info/whois/14.18.25.69
http://myip.ms/info/whois/118.85.207.18/k/3309037677/website/center.189.cn

Very suspect website traffic

Adding a mailing script and some logging to the 404 page of my website revealed the scary world of security hacking attempts, spammers, unwanted bots, old links and hopefully no missing files.

The Baidu spider started generating traffic going to URLs that do not even remotely exist on my site which in particular involves sign up, join and membership pages. This was later on followed by some three level deep news URLs and my website is only one level deep. Other bots started becoming an issue as well, but they did not keep going as long as Baidu did. At least Baidu is a big search engine, but the traffic makes me think that someone was using the search engines to scan my site to get access to sign up forms for spamming and other reasons. Since Baidu does not know of the URLs, it was most likely sending the bot to check if the URL is valid. If that is the case then I am happy that I was made aware of these attempts. They all got 301 redirects for their effort.

Another series of visits that I had was a blatant attempt at trying to breach security by looking for various standard updating URLs like WordPress and other CMS systems to try and identify what I use for my website plus possible breaking in. Due to custom system and security through obscurity, all of those attempts failed horribly and the 301 redirect list got extended. The CMS login is now also extended to three passwords and only three attempts to log in.

Another recent case that I had was URL's from other sites where "http://" gets stripped out and the rest gets added to maybe get lucky. My entire website is using full URLs and this was already happening when I used relative URLs.

The user agent string recorded for this was:
Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; MDDR; .NET4.0C; .NET4.0E; .NET CLR 1.1.4322; Tablet PC 2.0); 360Spider
IP Address: 101.226.166.243
HTTP Accept Language: zh-CN

Activate-A Mercedes-Benz tablet device app

Mercedes-Benz in South Africa has launched an app that work on iOS and Android tablet devices. It does not seem to be an option for entry level Android phones since I tried to get it. Part of the app requires checking in at certain locations. The app is supposed to show how the Mercedes-Benz A class car will look which is most likely going to be on certain locations like road, parking lot and others using augmented reality to experience that type of lifestyle. Currently I could not yet test this part due to having to get home first to install the app and then drive to these location.
To install the app, search for "activate-a" (without the quotes) in the app store. The app does require registration and works for Cape Town, Durban and Johannesburg areas.

Really weird website traffic from Baiduspider

In the past couple of days, my website got some really traffic from the Baiduspider.
This was all for pages that my website does not have which even included Yahoo news articles. My website currently only has one single news page and not a three level deep structure.
What became even more odd is this search for sign_up, signup, link-resources, register, join, member, membership and other files and folders like that. Not a single one of these requests was actually for a page on my website since I had everything else renamed for logging reasons.

The user agent of one of these page not found cases is:
Mozilla/5.0 (compatible; Baiduspider/2.0; +http://www.baidu.com/search/spider.html)