101 reputation
3
bio website
location Cambridge, United Kingdom
age 29
visits member for 3 years, 7 months
seen Aug 1 '13 at 23:54

I’m a bioinformatics PhD student at EMBL-EBI and the University of Cambridge but I’m originally from Berlin.

My programming interests span from C++ over .NET and dynamic languages all the way to XHMTL/CSS and R.

I’m mainly working on genomics using high-throughput sequencing data. My thesis is about the regulation and expression of non-coding RNA (especially tRNA) in mammals.

Here’s my …

twitter-pic Twitter account
g+-pic Google+ account
github-pic Github account
cv-pic Resume


Jul
29
comment What is the psychology behind laying out slabs at different angles for cyclists and pedestrians?
I’m with Kaz: when cycling I routinely drive onto the pedestrians’ half to avoid the in-line ridges.
Apr
16
comment Are age verification banners effective?
@DA01 This answer says that there is no legal requirement.
Feb
6
comment How can users be prevented from pouring water into the bean compartment of a coffee machine?
IMHO, this. If there’s a chance of confusion, the first idea shouldn’t be how to avoid confusion between two things, it should be to make the distinction unnecessary. If this solution should really prove impossible for technical reasons that cannot be overcome, then move to mitigating the confusion.
Dec
6
comment Can I safely remove search functionality?
Here’s a thought: what happens if you remove the navigation instead of the search? Though not an entirely serious suggestion, this would be an interesting UX experiment (provided the search is good). In fact, web search engines essentially operate that way (particularly compared to the previous ubiquitous web directories which have all but vanished). GitHub now also provides an (optional) search-only interface; smartphone interfaces lean the same way; and I believe that in the long run this kind of interface is the way forward.
Sep
25
comment Why does Apple use images of text for headers?
@JosephEarl Sure there are negative effects. Try selecting the elements and copying/pasting them. The alt attribute doesn’t prevent the fact that the headers are lost through this process. This is very bad (and in fact there are better image replacement techniques which mitigate this … e.g. the Phark method Matt mentioned).
Sep
24
comment Why does Apple use images of text for headers?
TypeKit’s most expensive business plan is 400.00 USD/month. Granted, that’s not cheap but probably not “prohibitively expensive” for an enterprise such as Apple. “Also, it makes the site faster with less to download” – did you benchmark? I’d think that the opposite is true for even a small number of header images, if the font integration is properly optimised.
Sep
24
comment Why does Apple use images of text for headers?
“not all browsers support font face” – well, to be honest all modern browsers do, and do it several versions back (since at least 2009 in the case of Opera, and before for all other browsers). And the Apple website never really catered to outdated browsers. They use HTML5 features such as canvas extensively, which wasn’t supported in MSIE until version 9.
Sep
11
comment Why this timed-hover “like button”?
I’d be wary of an interface that needs to give me updated instructions to be used correctly (the “don’t move”).
Jul
13
comment Should the space between the checkbox and label be clickable?
Let me phrase it this way: if I am filling out a questionnaire online and the space is not clickable, I curse the developer under my breath and close the questionnaire immediately.
Jul
10
awarded  Commentator
Jul
10
comment Why don't users want to see adverts relating to products/services they might be interested in?
If you want a source for your claim that ads are all lies: Feynman said so, and cited some cases on multiple occasions. Amongst others: in The Pleasure of Finding Things Out (can’t find the chapter, it should be online somewhere). He claims that advertisement is fundamentally intellectually dishonest and consequently unethical. Of course, Feynman didn’t survey whether people agreed with him but it would be a start.
Jun
26
comment Drawing users' attention to unrecoverable actions?
“OK” as a synonym for “launch the nukes” sounds too innocuous. People often don’t read the text, just the buttons.
Jun
26
comment Drawing users' attention to unrecoverable actions?
This is really the only acceptable alternative and should be used much more. However, I would definitely make the prompt more specific where possible – e.g. the file name to delete, as you mentioned. This prevents the user from just typing “DELETE” on auto-pilot without checking that they did delete the right file etc.
Mar
27
comment What type of graph do users prefer?
Statisticians generally advise strongly against pie charts since they distort the subjective assessment of the data with little benefit. In scientific literature, they are almost completely unheard of. However, if the chart only needs to offer an approximate comparison of different sizes then a pie chart may be appropriate, especially since it allows the viewer to put different sizes into relation very quickly.
Sep
7
comment What would be a good non-religious icon to represent that a person has died?
Try something new: use a refresh-icon; at least it will appease the Buddhists.
Aug
23
comment What's the best way to handle deleted user accounts on a social app?
This also depends on the legislation: in Germany for instance, if you don’t offer a user to irrevocably delete their contents, this will land you in a ton of trouble.
May
18
comment What effect would a grayscale color scheme have on UX?
Oh wow, I’d never seen the Physics.SE layout before. FWIW, it looked utterly, mind-blowingly amazing. But readability was indeed problematic.
Mar
17
comment What's a clean way to implement two checkboxes that can be “ANDed” or “ORed”?
If anything this is more confusing than the alternatives …
Feb
3
awarded  Autobiographer
Feb
3
awarded  Supporter