Blog

  • Upcycle Windows 7

    Upcycle Windows 7 : Microsoft announced support for Windows 7 would end Janury 14, 2020 with plenty of lead time. A regular procedure in the world of enterprise software. The idea behind such a process is fairly simple. The software won’t stop working, nor are users unable to use in any other way. Only the company will stop developing and supporting patches for the operating system.

    Upcycling Windows 7
    Upcycling Windows 7

    While this is typically not a immediate issue for the private user, it has some security implications. The corporate user, that requires support, still has an opportunity to pursue a more recent version of the software, Windows 10. The entire procedure created some media echo recently, given the date is only past due by one week.

    Instead of simply letting go, the Free Software Foundation started a campaign and petition to create an alternative for Windows 7 to just stop it. While this didn’t happen with software are recent as Windows 7, the approach has been precedented. MS DOS, Classic Word and even calc.exe are up on the internet nowadays. The Register mentions potential issues with content licensed from third parties, too.

    The announcement and petition is here: Upcycle Windows 7 — Free Software Foundation — working together for free software

  • Automatisierte Gesichtserkennung: Warum die Technik gefährlich ist.

    Automatisierte Gesichtserkennung: diese Woche is eine Recherche zu dem US Unternehmen Clearview durch die Medien gereicht worden. Die Artikel haben jeweils viel Aufmerksamkeit auf sich gezogen.

    Claudia Orsini. cyber. - Automatisierte Gesichtserkennung.
    Claudia Orsini. cyber. CC-BY-2.0, Flickr.

    Gesichtserkennung ist das Feature, das man bereits von Fotoverarbeitungsprogrammen, Handys und sogar Sozialen Medien kennt. Für die meisten Nutzer dieser Programme oder Dienste ist das meist ein lustiges, manchmal sogar nützliches Feature. In einer großen Sammlung von Fotos schnell alle für die Geburtstagsfeier alle Freunde wiederzufinden, ist schon praktisch.

    Wenn diese Funktionalität die Grenzen der privaten Nutzung überschreitet beherbergt die Anwendung große Gefahren. Zum einen handelt es sich hier im eine private Firma. Weder weiss eine betroffene Person, ob Ihr Bild in der Datenbank geführt wird, noch ist eine Kundenliste der Firma bekannt. Das bedeutet, dass die Anwendung der Datenbank ebenso unklar ist, und damit auch Missbrauchspotential eröffnet. So gab es bereits Fälle, in denen Beamte Frauen nachstellten. Eine Fotodatenbank erleichtert solche Vorhaben. Genauso, wie Regierungen und offizielle Stellen beispielsweise Videoüberwachung öffentlicher Plätze leichter auswerten können. Damit wäre denkbar, Bewegungsprofile von beispielsweise Regierungskritischen Bürgern zu erstellen.

    Die Technologie bringt durch Ihren Einsatz im öffentlichen Raum eine automatisierte Verletzung von Privatsphäre der Bürgerinnen und Bürger mit sich, wie auch Ulrich Kelber, Bundesdatenschutzbeauftragter, sich äussert.

    In der Süddeutschen hat Simon Hurzt bereits am Dienstag eine handliche Übersicht über die Problematik mit der Technik veröffentlicht. In dem Artikel sind die wichtigsten Fragestellungen über automatisierte Gesichtserkennung antwortet.

  • People recognition

    Only days after Clearview hit the news, The Economist runs an article on how the Combating Terrorism Technical Support Office (CTTSO), an agency of the US defence department, has developed another dystopian tool to identify people by their heartbeat.

  • The Y2038 Problem

    The Y2038 problem is similar to the Y2K problem. We’re exactly in between both about now. Both are 18 years away, in either direction. While Y2K is over and was obvious to everyone, Y2038 is not.

    The issue here relates to a representation of date and time in Unix systems, and is therefore sometimes referred to as Unix Y2K. The root is the convention to store date and time information as 32bit unsigned integer in such systems. This means, possible values are limited. Time-differences in seconds, starting from 01.Jan 1970 cannot span beyond 03:14:07 UTC on 19 January 2038.

    The Y2038 Problem
    Calendar

    The Y2038 problem will make all calculations beyond this date impossible, until migrated to another representation. At the time being, this seems far away. However, the problem casts its shadows already. Industries, in particular financial markets, often rely on long term forecasts.

    Governance issued treasury bonds come with with the longest maturity. Often twenty years, sometimes thirty years. Calculations for complex, long running financing models easily try to estimate returns 20 years and beyond into the future. This is already beyond the problematic date that Y2038 brings. The code to run these calculations is typically complex and stable. Sometimes, it is as old as from 1970. Back then, this date-representation Unix engineers introduced this approach. 32bit covered a long period. John Femellia has a thread, over at Twitter, telling a story about the upcoming issues today.

  • This week in dystopia.

    This week in dystopia: The New York Times has an article about the next steps in dystopian future. A start-up evolving face recognition algorithms, fed by a database with facial images, scraped from the open web.

    Clearview - This week in Dystopia.
    Clearview – This week in Dystopia.

    A little-known start-up helps law enforcement match photos of unknown people to their online images — and “might lead to a dystopian future or something,” a backer says.

    The New York Times: The Secretive Company That Might End Privacy as We Know It

    Further, the article describes the sheer size of the database. At a rate of massive duplicate numbers, three billion images is still impressive.

    The system — whose backbone is a database of more than three billion images that Clearview claims to have scraped from Facebook, YouTube, Venmo and millions of other websites — goes far beyond anything ever constructed by the United States government or Silicon Valley giants.

    The New York Times: The Secretive Company That Might End Privacy as We Know It

    In times in which criticism of big tech is on the rise. Just this week Jannis Brühl, Head of Tech News Department at @sueddeutsche Zeitung published an opinion that this technology is dangerous and should be banned,. The article include an appeal to German government to create legislation to do so. Jannis is in good company with other tech critics like Eyvgen Morozov

    Source: The Secretive Company That Might End Privacy as We Know It

  • Product Owner vs. Product Manager

    Product Owner vs. Product Manager: Product Management is a challenging role and requires diverse skills. Large organisation often introduce a split between two similar, close roles – Product Ownership and Product Management. Both requires a large set of skills.

    Jordan Bergtraum, The Product Mentor, a mentor at The Product Guy, leads a conversation on this split.

    Source: The Product Guy.

  • Microsofts Patch Tuesday

    Y’all install Microsoft Patch Tuesday patches within 24h, right? This time Krebs On Security has some rumours that make you want to really install these patches in time.

    Update: Washington Post reports, the NSA warned Microsoft about it.

  • Hacking Machine Learning

    Hacking Machine Learning

    For fun and profit.

    Could still be bad coding practice.

  • British Royal Mail issues computer games stamps

    British Royal Mail issues computer games stamps writes BoingBoing:

    BoingBoing.net
    Worms, Wipeout, Micromachines & Sensible Soccer

    The article features pictures of these stamps. Those who remember the times will find these adorable. The stamps are apparently available as first or second class stamps, at £1.60 or £1.55, and also in collectors sets.

    Do I know enough people in the UK to send me postcards with stamps from:

    • Elite(1984)
    • Dizzy (1987)
    • Populous (1989)
    • Lemmings (1991)
    • Micro Machines (1991)
    • Sensible Soccer (1992)
    • Wipeout (1995)
    • Worms (1995)
    Lemmings, Popoulus, Dizzy and Elite

    each? Germany, meanwhile, is still debating whether computer games are hazardous and the people involved in the scene should be observed.

  • The North Star Playbook

    The North Star

    As a concept, the North Star principle gained a lot of attention in Product Management recently. Amplitude, a vendor of analytics tools, has a guideline to this concept. Their playbook walks product managers, those that want to enter the domain or even those just curious about methods and principles through the ideas. But also sets the scenes for potential applications by walking through exemplary goals to achieve with this approach.

    Stapler - The North Star
    Stapler (by James Bowe)
    Picture unrelated.

    The playbook comes in 7 chapters, starting by describing the ideas to apply with the North Star concept. Only after the introduction the playbook enters the practical application of the concept, and with a chapter on product metric checklist checklists, it emphasises the importance of metrics. With this it also stresses the importance of selecting the right metric and not to lose a product in vain. E.g. active users would be the wrong metric, given the goal that shall be achieved.

    More practical guidelines come with the chapter on running workshops in part 3, and the chapter on defining the right guiding metrics. In between, the document also gives success stories: there are sections that talk through a successful implementation of North Star at Netflix. But also Amplitude is leveraging the methodology and shares their experience in a section.

    The closing chapters dedicate to debugging the processes attached, implementing them and over time changing directions.

    Takeaway

    In all the recent hype around the method, the key take away is to simplify ideas for your organisation. The approach is supposed to make it easy for your teams to understand the direction the product is taking. And even more following this direction. For a product management, communicating ideas should be a core skill. This approach gives great tools in doing so.

    The guide to discovering your product’s North Star to improve the way you manage and build products.

    from the Amplitude Playbook

    Source: AmplitudeThe North Star Playbook