Profile

Cover photo
Istvan Soos
163 followers|47,135 views
AboutPostsPhotos

Stream

Istvan Soos

Shared publicly  - 
 
My colleagues gave this talk about our work. Yay!
 
Had the opportunity to give the day two keynote for the Dart Summit this year.
2
Add a comment...

Istvan Soos

Shared publicly  - 
 
Science. It is always great to know how the world works.
 
Oh, we'll just slingshot it around a few planets to gain speed and rendezvous with the comet... easy.

Science, it works. As a bonus, check out the photos from Nasa: https://www.flickr.com/photos/europeanspaceagency/sets/72157638315605535/ 
1
Add a comment...

Istvan Soos

Shared publicly  - 
 
A must-see talk for history geeks about the early development of Silicon Valley.
And Ye Shall Know the Truth, and the Truth Shall Make You Free - John 8:32 (and the lobby of CIA Headquarters, Langley Virginia) A few other versions of the talk are here and here Secret History Ba...
2
Add a comment...

Istvan Soos

Shared publicly  - 
 
 
Ever wondered why we demand 60FPS? ... But most movies still record at 24FPS? +Paul Bakaus has the answer. 
A story about the perception of vision, frame and refresh rate, motion blur, and TV displays.
1
1
Roman Kamyk's profile photo
Add a comment...

Istvan Soos

Shared publicly  - 
 
Great resources on how to design for mobile, tablets and desktop web at the same time.
2
2
Roman Kamyk's profile photoAxel Winter's profile photo
Add a comment...

Istvan Soos

Shared publicly  - 
 
Performance and perceived latency matters. </confirmation bias>
 
Speed, Performance, and Human Perception - brief keynote talk I gave earlier today at Fluent conference.

TL;DR: optimize for task completion and learn about the (intriguing and often bizarre) properties of the wetwear sitting between our ears! There is more to performance than milliseconds, kilobytes, and frames per second.
1
Add a comment...
Have him in circles
163 people
Gallai István's profile photo
István Szabó's profile photo
Richard Walker's profile photo
Beáta Malek's profile photo
Ted Sander's profile photo
Balazs Molnar's profile photo
Zoltán Mózes's profile photo
Henry Bannister's profile photo
János Pál's profile photo

Istvan Soos

Shared publicly  - 
 
It all started as a very small disagreement on how to call certain fields, and it eventually grew into a reasonable naming guideline that I've always missed. Strongly recommended to read even if you are not working with Dart.
 
numPages or pageCount? toSomething() or convertToSomething()? The new API Naming Guide lays out the guidelines that Dart engineers use when choosing API names.

#dartlang
Choose good names to make your API easier to understand.
3
Add a comment...

Istvan Soos

Shared publicly  - 
 
One of the many reasons why front-end performance is hard to do right.
 
At first glance, HTML5/CSS seems to have amazing capabilities for next gen UIs: any of the 115 or so style-able properties can be animated. Yet, in modern browsers only two properties, transform, and opacity, can be mutated at 60fps. In this talk, I explain why thats the case, and ask the question, "Should we demand better?"
1
Add a comment...

Istvan Soos

Shared publicly  - 
 
Yay for +Paul Rohde and +Ted Sander! Big web applications, big problems, really great suggestions and solutions. And of course: #dart  language for the modern web.
5
3
Ted Sander's profile photoIstván Nagy's profile photoMark Bridge's profile photoAxel Winter's profile photo
2 comments
 
Some of it is... Of course if you are dealing with something deeper like say a crazy powerful Table.... That is definitely not a piece of cake ;-)
Add a comment...

Istvan Soos

Shared publicly  - 
 
As I'm working with Paul and Ted (*), I can truly recommend their presentation at Google I/O (Jun 25). They'll discuss large web apps, technologies and team dynamics, and our Dart work within AdWords. Watch them if you are interested in how to build large (enterprise) web applications (in Dart).

#dart   #googleio2014  

(*) Hey, they've just shot at me with a nerf gun. Better be on the safe side and check the other speakers too:
https://www.google.com/events/io/schedule
Changing the implementation of a business\u002Dcritical web app without any downtime or user interruptions can be nerve\u002Dracking. As software engineers, it\u0027s what we\u0027re asked to do every day. At Google, we know all about keeping our users happy and productive while swapping out low\u002Dlevel implementations. In this session, we share proven techniques for writing business\u002Dcritical web applications, no matter the size of the ap...
10
8
Guillaume Girou's profile photoDan Grove's profile photo
Add a comment...

Istvan Soos

Shared publicly  - 
 
Agile project management guidelines in government projects (UK). Details are missing, but even the consideration and effort that was put into this guide is amazing (considering the inherently waterfall-like nature of gov projects).
To help with efficient planning and approval of spending proposals for agile digital and IT projects, the following clarification of business case guidance has been produced in collaboration between HM Treasury and the Government Digital Service. This should release the potential of the agile ...
5
2
Rafal Gasiorek's profile photoSavan Vyas's profile photo
Add a comment...

Istvan Soos

Shared publicly  - 
 
This is a must-read for dealing with large, realtime datasets with event sourcing, from the creator of Storm, a distributed realtime computation system[1].

"What makes scalable data systems difficult isn't the CAP theorem. It's a reliance on incremental algorithms and mutable state that leads to complexity in our systems."

"I turned CRUD into CR, split persistence into separate batch and realtime systems, and obsessed over the importance of human fault-tolerance."

"Mutability is really just an inflexible form of garbage collection (that also interacts poorly with the CAP theorem)."

[1] https://github.com/nathanmarz/storm/wiki
#cap   #eventsourcing   #bigdata  
The CAP theorem states a database cannot guarantee consistency, availability, and partition-tolera...
1
Roman Kamyk's profile photoIstvan Soos's profile photo
2 comments
 
Well, the title's "beat" part might be a bit overstatement. My understanding is that the window of eventual consistency can be kept short with his method, depending on how the 'realtime' part can be implemented. The fact that the  implementation remains simple is a huge long-term win.
Add a comment...