יום חמישי, 27 בדצמבר 2018

Debriefing in the era of Agile Software Development

Software Development has been traditionally managed in the Waterfall model or spiral/iterative models. In the Waterfall model, each stage of the development had to be completed before the next stage could be initiated.
First Design, then development, then testing. At the end of each stage, an evaluation process had been conducted. If needed, the process would go back to previous stage to fix the issues.
These evaluation points were kind of debriefings and reviews for what happened up to that point.
However,  the lengthy stages were met with dissatisfaction from stakeholders. When the first products of development arrive, after months, the stakeholders would have had opinions for re-design.

With the spread of other iterative methods, the Agile method got its grasp on the high-tech industry. The idea behind the Agile development process is that "control points" are set and fixed all along the process in short-time periods.
This way, stakeholders can provide feedback and the developers or the people who work on the product can respond more quickly, with more agility. One of the most popular implementations of Agile is the Scrum method.
I won't go into full detail of  the Scrum method but I'll focus on the ceremonies. Each team that implements Scrum conducts some ceremonies that assist with planning, daily review of status and retrospective meetings.
At the end of each development sprint, which usually takes 2 weeks and includes design, planning, developing and testing of small subset of the product specs, the retrospective ceremony takes place. In this ceremony, the team members perform a debriefing of the past 2 weeks:
- What went well?
- What went wrong?

The major focus is on the team's work, The review does take into account external influences ("the deign team made a lot of errors, the DevOps didn't give us enough servers"). But, the real deal is reviewing the team's internal work, co-workers communication, adherence to procedures, etc.
Team members are asked to raise problems and solutions.

The retrospective is a crucial part of the process and without it, same issues and problem re-occur again and again. In IBM XIV GUI team, at stressful times, such as prior to product launch, some of the scrum ceremonies were postponed in order to give people all the time they need to work.
However, we would have never skipped the retrospective, especially in stressful times. This debrief is important in keeping the team's work efficient and to allow quick solutions for one-time and recurring issues.

יום שישי, 21 בדצמבר 2018

Communities in Practice and Meet Ups

The latest topic of out course, Communities of Practice got me to think on the influence of social media, and the internet as a whole, on the way communities may interact.

One would expect that with the growth of social media and social networks, people will use those platforms to collaborate more, rather than actually meeting and socializing.

Facebook, Twitter, form platforms and other platforms promoted creation and fortification of communities of practice. In recent year, several websites that specialize in promoting community gatherings popped up: mainly meetup.com, everbrite, Facebook events and more.

Those meetings and gatherings complement the online platform in ways that let individuals to communicate directly without any means. One can converse with many other people easily by speaking, as opposed to typing and reading which is a heavier cognitive task.

So even though online social platforms play a major role in bridging different people from different geo-locations, face-to-face meetings are here to stay and both "platforms" can benefit from one another.

יום חמישי, 13 בדצמבר 2018

Designing Social Spaces - Reputation and Gamification

One of the challenges in creating a successful and lively social platforms, is the question of continuously engaging users.

It is obvious today that knowledge management is based on the social atmosphere of companies and organizations. Do we want to hand out bonuses? Maybe deduct salary from users that don't contribute?

One approach is to use reputation. As human beings, we have a strong sense of community hierarchy and respect towards one another. People that have higher reputation are honored with higher respect and taken more seriously. For Jewish religious people, there are different levels of Rabbais. Same thing goes for professors or other high ranked personas.

Stack overflow, the most popular software development Q&A platform, has a sophisticated reputation system. Every action a user performs or interaction they're involved in produces an increase in reputation: answering questions, being up-voted on your answer, leaving a reply, up-voting answer etc.
This methodology proves to work because many users invest a lot of time in maintaining their reputation by being active on the platform. People with high reputation are considered to have better answers. It is so successful that companies who seek to hire employees look into their candidates ranking in Stack Overflow. Some candidates even include it in their CV!

Gamification is a bit more advanced approach. While it is kind of reputation ranking, since there's an explicit or implicit ranking, it is mainly focuses on adding fun to the intended task.
We can look at Waze or Moovit as socially-powered apps. At their begging, both apps relied on crowd-sourced updates for fixing map data, road data, incidents report, real-time reports and more. Both companies used games to encourage users to report. Waze encouraged users to drive through certain locations to improve their database by adding virtual tokens to their balance if they pass there.

These tools can encourage users to share knowledge and insights, but they have to be crafted carefully to be successful (Waze & Moovit dropped their gamification incentives as time went by...)

Has Wiki transformed the Knowledge Management Culture in Organizations?

Until 2010, the most common way of managing knowledge in enterprise organizations was Microsoft Sharepoint websites.
Users would create documents and upload them to the Sharepoint. The novelty of the Sharepoint was a useful and convenient way to share documents between users with versioning. If once had to have shared volumes with folders inside where one user might overwrite changes, now Sharepoint allows version control and recovery.
However, Sharepoint still consists of documents as the smallest component of knowledge. The user has to checkout the entire document, edit it, and check in again. Other users can't edit it simultaneously and if the document wasn't check out, merge conflicts occur.
These are exactly the kind of issues that prevent user from documenting and contributing their knowledge to the organization.

Then, came Wiki. Mainly by popular spread of Atalassian Wiki, users can now contribute their knowledge by directly editing articles in their browsers. The knowledge and data aren't stored in documents, but in online editable articles that in later versions, user can even edit collaboratively.

So, Wiki tools do make it easier for users to contribute data and edit articles. Nonetheless, as with all tech tools, the ease of updating wiki resources still need to be governed by KM policy. Companies always have to get their employees the easiest and best KM tools, but to enforce KM policies by compulsory, guidelines, bonuses etc.

יום שישי, 23 בנובמבר 2018

From payslip to personal crowd-funding

During yesterday's class we discussed the issue of crowd-sourcing - how can an organization, company, or any other entity can leverage the power of other people in fulfilling their objectives.
If we look at this term with a broad view, crowd-sourcing can take many different forms:
- P2P micro loans that allow people who can't access banking services to start up their own business
- A company that out-sources certain tasks to many individuals. The collaborate work is fast and efficient in this manner.
- Leisure and dining review websites that present crowd opinions and more.

Recently, I have come to learn about a new type of crowd-sourcing: direct customer funding by the Patreon platform. 
In general, people who work are either employed by a company or act as free-lancers offering their services to individuals/companies. In this sense, a free-lancer needs to offer some sort of product or service to an individual. The individual than pays for that long-term service or product.

With the evolving of internet and social platforms, people have started offering their services, wisdom, creations, experience online for anyone to consume. 
Take me for example - I'm a software developer in intermediate level. My latest desire is to combine my interest in data vitalization development on the web using  D3.JS with the newest tech available for web development -  React JS. The two techs don't play well together out-of-the-box.
This one blogger, Swizec Teller, has started writing a book and numerous posts about combining these two together. Swizec mentions that he gained some insights about combining these two technologies from other persons, but mostly he's done alot of research and trail-and-error attempts. 

Thanks to Swizec, I'm able to learn new skills. However, while it might take time to Swizec to publish a book, I think he should be credited for his efforts. He puts alot of effort into his work.
Patreon allows me to back him up by, let's say, 5$ a month. 
Combined with other people direct back-ups, Swizec can put more effort into his work which in turn benefits everyone else with new knowledge and skills.
Apart form techy people, artists such as designers, musicians, authors and others can establish a successful crowd base that can back up their publicly available efforts.

Hopefully, with enough backers, Swizec can be less dependent on his payslip and more on his crowd of back-uppers.

יום חמישי, 8 בנובמבר 2018

After 15 years, does IT still don't matter?


Back in 2003, Nicholas Carr published his article “IT Doesn’t Matter” where he states that as IT become more ubiquitous and easy to consume, they become more invisible. During today’s class, we discussed the question whether a CEO needs to address IT the same as the electric network, as an example of a commodity service.
Today, following 100-150 years of development, electricity is consumed as a service – a person, or a company, contacts the electricity company and gets connected. This is thanks to years of practice that got us to a point that electric lines and infrastructure are taken for granted in new buildings and considered a must-have everywhere. So, a CEO doesn’t need to concern herself with the way the power is supplied, as long as it does in a reasonable price.
If we cast this analogy on the IT era, even though it’s been 15 years since his publication, I don’t think that Mr. Carr is quite right (yet). If we look at IT companies, such as software companies, mobile app developers and other, the CEO must be involved in IT processes as these are the main product or service produced by her company.
What about other companies that are not focused around IT? 
Well, the world is still in a transition towards a 100% IT-as-a-service. Let’s leave aside the possibility that we might get there, we can see that some services are already provided as a service while others don’t. Any company can get computation and storage services using AWS. In the past, the IT manager would have to get the CEO’s approval for server equipment and other high-detail decisions. But with AWS and similar services, the “rails” that run the service are not interesting – only the price and quality. The risk is lower for a waste of budget, as the physical infrastructure risks are out-sourced to Amazon.
And yet, CEOs are expected to still be involved in IT-related decision making that might have strategic implications. Shufersal, a major food retail company in Israel has been trying to go online for more than a decade. The company has an online supermarket website, developed solely for the company. The website is a bit old-fashioned, but was developed as one of the first websites of its type in Israel:
                                                

Recently, Shufersal announced it’s going head-to-head with the upcoming entrance of Amazon retail services in Israel. This decision includes launching of a new website that will offer commerce options that do not rely on food products, but general commerce including imports. This is a strategic decision that requires a huge IT effort and investment – it’s unique for Shufersal. This is the upcoming website current status:
                                              
The CEO must be involved in the details of the effort required for this kind of transition.
On the other hand, other food retail companies in Israel are not interested in going head-to-head with Amazon. Tiv Taam and Mahsanei Hashuk apparently want to focus on their current domain. If you go to the online version of supermarkets that belong to these 2 companies you immediately see they have a really similar user interface:
                                                             
The companies are totally different with different stake holders – we can only conclude that their online supermarket is a product of an online service that allows food retail companies to out-source this need.
So in this case, the investment and risk in establishing an IT support for online retail is probably lower than Shufersal in past days and now.
CEOs cannot still treat all IT aspects as a service, but we’re slowly going there.

יום חמישי, 1 בנובמבר 2018

Fake News - now and in the 90s

My name is Shaked Kaufman, a Master’s student in Industrial Eng., Ben Gurion University.
This blog is hopefully going to allow me to express my views and opinions about Knowledge Management theory and practices as part of an academic course.

It’s interesting to see the role of “garbage” data in our daily life and how it may influence it greatly.
The term “Fake news” wasn’t coined recently, but it became really popular during the 2016 US presidential elections. Looking at Google Trends (https://g.co/trends/VScNw) we can see that since 2004 the use of this term was pretty sparse, but a surge took place at 2016.

Take The Onion website as a great example for actual fake news. Since 2007 (and even before as a printed edition), the website publishes satiric articles about a wide range of international, local, cultural, political and other issues. All the articles are humor and convey a distorted and sarcastic view of reality. 
What makes The Onion unique, compared to other sources of amusement and humor, is that its articles look genuine as if they were actual news articles. For the casual Joe that bumps into one of their pieces, this might look like very odd news.

Myself, I failed for some time to understand why עמ;לק summarizes such strange articles!
As it dawned on me, there’s a cool service called אמ;לק, the equivalent of TL;DR. The service allows readers to summarize posts and articles for the benefit of other users that want to get the bottom line.
עמ;לק is a satirical service, however, that with a single change of a letter entitled itself with a sophisticated name in Hebrew (referring to an biblical arch-enemy nation of the Israelites). It summarizes real articles in a humor and totally incorrect way.
So no harm there - after few posts I understood my mistake and confusion.

And while sites or services like עמ;לק or The Onion are only meant for humor and fun, in 2016 we learnt that Fake News can be very not amusing and also very dangerous.

Allegedly, it has been reported that Russia intertwined in the US elections using fake posts and news generated in order to influence on the American people opinions and views.
In other cases, men were accused in India for rape or other criminal acts in a WhatsApp message. The messages spread quickly and an angry mob killed those. Whether or not these men actually committed the crimes - they were taken to drumhead court without proper justice measures.

Both the tech companies and the society should address the issues of fake news and the ease which they can be conveyed without cross checks.