Category Archives: IT vNext

Where do they go, once you let them go through your port (number)?

As you know, UDP is stateless (or to called connectionless). So there is no guaranteed delivery to destination (nor promise what being returned to you).

UDP grow rapidly in popularity last years. Common known areas is online gaming, media streaming, online collaboration and even some data intensive business applications rely on UDP. It’s also nice to mention services like Wifi loves UDP to distribute discovery information to the network.

However, with UDP came critical levels of security risks. Beside that you, by protocol definition. don’t know for sure if the packets is delivered (or not), you also send information that are capable to redirect, transform or drop without further notice. When they return, you think you know what come back – but don’t know for sure.

So the retoric question: Do you care or know in detail what data you provide in UDP packages? Once the data is kept in a encrypted transfer, is fine, as long as the encryption is secured.

Happy weekend 🙂


This posting is sent to remind about not produce exceptions for flow control!

Why? Because the code will be hard to follow. One path for the exception and one without.

Why? Because the exception may not be good for what’s really happen.

Why? It might be re-written without knowledge of what happen in the methods with the exception control flow

Why? Because you would probably not reverse trace all instances of your exception

Why? Because you are lazy

Why? Because you know that you won’t reimplement the methods.

Why? Because you find out it’s not within the scope of your job.

Why? Because the boss will ask what you are doing instead of the described change task.

And by the way, maybe the automated tests or other mechanisms do, because they might rely on the exception flow.

And then.. then you have to ask the boss why tests failed when you worked on something else.

– Just do the the exception handling properly! Normalize the collections based on agreement of data structure (or on the lack of it) during increment. Create specialist classes for anomalies and exceptions, such as it was new programs to handle the code.

Data science evolution 1960 to 2040

Isn’t it amusing to find out how paradigms have passed in the computer era? Let me share some weekend amateur drawings.

1) 1960-1980. Monstrous servers and ultra dumb clients.


2) 1980-2000. Monstrous clients and pity servers connecting to outputs. (VB6 and access databases loved to be close to the client).

3) 2000-2020. Enormous servers and enormous javascript heavy clients exchange enormous amount of data everywhere. What’s a paradigm here?

4) 2020-2040 (speculative). Connected clients is units, owner of it’s own data, stored locally. Data transfer between units and a kind of interbackbone is just descriptions, models, meta, statuses, point-to-point-connectivity agreements for subscribers. Interbackbone is just there to transport data, act services, agreeements and do physical transfer between networks. More like IP dgrams in the lower layers.

Have a nice friday and weekend! And remember to do your homework.

The florist steals your data

I planned to steal information based on possibilities on my current flower care program and personal interest to earn money and give the rich world side effects of greed. Yeah – plant flowers. Not really my home genre but why would it be important. You would be surprised how less people care about flowers in the office. Let’s concentrate on two very interesting customers of mine. Five visits a year on this traditional big bank, let’s call it Sach. Three visits a year on another new coming popular Internet based bank, lets call it Prls. How hard can it be. Luckily nobody recognize that I work for both of them, even but the agreement is in the third year now.

This third year was about to be special. After this year in total eight visits, I have added a dousin of sensors in different areas of the offices that I care. They together collect approximately 1GB of data per 24h. Already in mid of second year agreement, I proposed a replanting of flowers next year. More specific, the 2nd visit at the large bank Sach and 1st visit at the Internet bank Prls.

In parallell I already tried a personal study of the planting. I had a year to find out that the flowers will survive. When the time for replanting occur at the customer, I have full responsibility over everything from bring flowers to- and from the office and including the potting soil to use. Nobody watching me and could care less of how I do my work. So except planting, I did also add  two water-resistence battery bays and a wireless hotspot in the pots. Special manufactured hardware and sotware for this purpose, that broadcast on a hidden network on another range. From the batteries I pulled thin wires inside the thicker stems of the bigger flower, pluggable cords inserted from one of the lower branches. The smaller flowers can’t use the wires but can collect sound and movements.

Next visit after the replanting, nurture and prune the flowers, I not only do this. I also verify connectivity and read the so far collected data into my smart phone, during the time i spent on each plant. I also punch a microcamera into the plug spare below the bottom bransch. Just 3 millimeter in size but with quality just as good as I can do face and basic image recognizition. Even smaller sound recorders in all flowers and not least, sensors that register possible movements. Next visit I will be able to transfer approx 100GB of data per flower, that will be stored since my last visit. A success transfer will take about 15 minutes and automatically clean the memory card once done. I walk on to next big flower. The smaller flowers does collect sound but also information about other networks in the office.

After each visit, I sell the data unstructered on dark web on auction. The buyers get exclusive rights (Well they can of course steal it if they want, but we trust each other) to re-distribute as long as they return to me with useful information they extract. The pay is mostly BitCoin and unfornately I know almost 100% black market and worser. In turn I use the money to pay intelligent developers create nice techniques and software algorithms for me. Why would I have moral panic for providing information from organizations that more or less steal money from tax paying people, instead of show a truly interest to reinvest them on a better world. As soon as more legal money is reinvested to the people and a better world, I will stop steal and distribute the information. Last time something really useful happened, was when some gigabyte of mail server information was exploiting some tax-free accounts and related rubbish coooperation.

What they do with the data? I can imagine. Myself is running automated pattern analysis so I identify for instance faces visiting pattern. I know for instance how many cup of coffees or visiting time in a particular toilet customers have. Funny enough the algorithm find that a higher percentiles of visiting time is incredibly higher at about 15 o clock until office close on afternoon, then the rest of working day. Algorithms do also set placeholders on media where there are more then one voice involved, but just one a time talking – appear as a chatting. Again, the coffee machine bring many discussions and facts.

Looking back

It begun back at 2013 when i was unemployed, i met a guy at a bar. I was alone at a round table with space for four in a rounded shaped red leather sofa. In front of me, except some other tables and the bar, I had a book with the title “10 paradoxes in human behaviors” on top of another book with a very basic title “Computer Science” on the table. At that very moment I also had a boring Pang IPA beer on my table. For me, nothing was unusual with this except that time I met this guy.

He asked to sit with me for a while, having some questions around my choice of books. Later he asked me about my daily life, and then tell me a little about his. I lie a lot to him, described my work as a buyer on a book store nearby. I read a lot of books and have an idea about what people need to hear for me to be trustworthy. This way we started to know each other and he proposed me a job as a florist, to work at offices to take care of their flowers.
A maked up story by Dundee

Hur ska IT-branchen bemanna och finansiera de drömlikt stora projekten?

LĂ„t oss starta med denna parabel: Anta att byggsektorn fĂ„r ett projekt, ex förbifart Stockholm, Öresundsbron, Nya karolinska, Inlandsbanan, Citypendeln och sĂ„ vidare. AlltsĂ„ riktigt stora projekt. SĂ„dana som tar nĂ„gra Ă„r att slutföra. Man mobiliserar en viss omfattning av arbetskraft för en tid.

LĂ„t nu sĂ€ga att vi mobiliserar samma omfattning av manstimmar och projekt för att bygga nĂ„got inom IT. Vad skulle vi bygga dĂ„? Du kanske har fler idĂ©er Ă€n mig, men för mig Ă€r tanken lite svindlande. Kanske man kunde genomföra omkoppling av den vĂ€rme som enorma datacenters generar, till att vĂ€rma upp bostĂ€der och kanske hela markareal? En redan vĂ€l diskuterad idĂ© och redan Ă€r i ropet pĂ„ flera hĂ„ll. Jag tycker man kan tillĂ„tas fantisera och överdriva. Det Ă€r ju vad som emellanĂ„t sker i andra byggprojekt. Man har nog ocksĂ„ skrattat Ă„t tanken pĂ„ att bygga snötunnlar i den Arabiska öknen, byggnader som Burj eller varför inte en tunnel mellan Dover och Calais. En konstruktion som Öresundsbron byggs inte sĂ„ ofta. Inte heller Högakusten-bron. Inte heller Turning Torso. Tellus Tower byggs ju inte heller sĂ„ ofta (knappt alls, dock).

Men nÀr och hur ofta byggs ett högprestige IT projekt? Var Àr visionerna hÀr? Varför verkar de IT-projekt som hittas pÄ sÄ svÄra och krÄngliga? Diffusa och höga risker avseende levererbarhet och pengar? NÄgon har kanske redan snudda vid tanken att det inte gÄr att fÄ tag pÄ resurser att genomföra enorma IT-projekt. Kanske nÄgon hunnit börja skratta Ät kostnaderna det skulle innebÀra. Att bygga nÄgot riktigt stort inom IT, hur skulle det betala sig? Kanske ocksÄ den som Àr verklighetsmedveten och jordnÀra ser ett ökande hot hÀr. Att IT kanske halkar efter. Det gÄr inte att finansiera eller ens genomföra ett stort projekt i landet? Skulle man kunna mobilisera ett slags IT försvar, ponera att de pÄgÄende spÄren av informationskrig tilltar?

Bygg ut – fram för dussinyrken inom IT

Det Àr min fasta övertygelse att IT Àr nog moget att ta in ett större omfattning av dussinyrken och medelmÄttigt utbildade och kunniga medarbetare. Normalutbildade eller inte alls (pÄ omrÄdet) utbildade som pÄ en introduktions- och utbildningstid kan ta sig an en uppgift, utföra den med bravur och fÄ en normal lön. Om och om igen. Ett nytt verkstadsgolv, industrigolv eller stora styrkor trÀnade för ett ÀndamÄl. IT av idag Àr nÄgon sorts överskattad bransch nÀr det kommer till kompetens. Det fungerar inte att ha bara specialister och generalister. Att utbilda sig inom IT idag handlar om att sÄ fort som möjligt kunna allt och med rÄdande kultur ska man snabbt avancera och krÀva löner för det minsta man Ästadkommit. Men det behövs automatiseringar, kvalitétstester, stickprov och inte minst repetetitivt mÀngdarbete.

Precis som gruvdriften kunde fixa till redan pĂ„ 1600-1700 talet? LĂ„t mig dra en parallell pĂ„ LM Ericsson. LĂ„t mig ocksĂ„ prata i dĂ„tid, eftersom det Ă€r dĂ„tid för mig. SjĂ€lvklart Ă€r det hĂ€r ocksĂ„ levande nutid i vilken tillverkningsindustri som helst. Testare, felsökare och montörer kunde tas in i princip frĂ„n gatan. Det togs Ă€ven in före detta undersköterskor och sjuksköterskor i stor omfattning (eftersom det blev bristyrke och dĂ„ligt betalt under en tid, sĂ„ ett antal lockades till industrin). De började med kort intro utföra arbetet med bravur. Felsökningar, reparationer och utvecklande av nya produkter krĂ€vde dĂ€remot det lilla extra. Ingengörer och konstruktörer, men sen erfarna och duktiga arbetare “pĂ„ golvet”. Internutbildningar och nĂ„gra Ă„rs erfarenhet rĂ€ckte för att kunna sticka ut med kompetens för detta. Exvis för att programmera. Ingenjörerna fanns (finns!) dĂ€r att justera och fixa nĂ€r nĂ„got föll utanför automatiseringen.

IT en yuppietrend – en ny slags golf

Visst Ă€r det spĂ€nnande att se att IT yrken generellt stĂ€llt sig högt i kurs. KarriĂ€rister lockas hit och det Ă€r lite spĂ„r av en “ny slags golf”. Folk med allt högre profiler söker sig hit, IT bolagen söker ocksĂ„ folk med högre profiler. Det kommer löner, bonusar och förmĂ„ner som börjar likna nivĂ„er som annars mer kopplas till prestige. Inte helt sĂ€llan hörs ocksĂ„ att IT stĂ€ller synligt i grannskapet bland det annars “glammiga” Finans. Jag skulle ocksĂ„ vilja snudda vid det löne-, förmĂ„ns- och bonusrally som pĂ„gĂ„r sedan mĂ„nga Ă„r, mellan och inom de som etablerat sig i IT-sektorn. Ett rally som hĂ„ller högre profil Ă€n tillströmningen av kompetens. Är det nu inte dags för IT att börja rekrytera de tyngre korten frĂ„n vĂ€l inoljade tillverkningsindustrier? VĂ€lkommen hit, Volvo, ABB, Skanska och sĂ„ vidare! Bort med tjĂ€nstesektor-essen! IT behöver dussinyrken nu, dĂ€r man snabbt kan rekrytera och bemanna i mĂ„ngfald!

Börja transformera!

  • Ta Academic Work pionjĂ€rsanda med mĂ„nadskurser pĂ„ allvar: Ja, det gĂ„r att göra IT konsulter pĂ„ nĂ„gra veckor. Men gör den Ă€nnu lite mer effektiv. Ännu smalare resurser och Ă€nnu fler.
  • Ta in profiler frĂ„n befintliga och erkĂ€nda tillverkningsindustrier och verkstĂ€der med god vana av att mobilisera hundratals och tusentals anstĂ€llda. Varför inte ta en titt pĂ„ vĂ€rnplikten.
  • Arbetsplatser med dess geografiska placering inte Ă€r central lĂ€ngre. Utvecklandet eller testandet behöver inte ske dĂ€r kunden sitter. Det har sagts sĂ„ mĂ„nga gĂ„nger, men tĂ€nk Ericsson Cables. Byggde man kablarna dĂ€r de senare skulle dras, eller skickade man dem pĂ„ lastbil?
  • Sluta bemanna projekten med resurser och personal sĂ„ seniora att de snudd pĂ„ kan driva hela projekten sjĂ€lva. LĂ„t arbetsbeskrivningen för dessa starka förmĂ„gor handla om att bli ansvariga för att leda och transformera deras omrĂ„de till att passa för dussinbemanning.
  • SlĂ€ng ut affĂ€rsledarna ur IT Projekten. De kan driva sitt affĂ€rsintresse, men inte det tekniska projektet. Placera folk som kĂ€nner till verkligheten och kan sĂ€tta ned nĂ€ven i styrelse/ledarmötena om vad som faktiskt krĂ€vs pĂ„ de olika idĂ©erna.
  • LĂ„t oss komma ifrĂ„n löst sammansatta projektgrupper dĂ€r 50% av kompetenserna överlappar. Grupperingar dĂ€r kanske en tredjedel kĂ€nner sig i starkt behov av att avancera och tĂ€nka pĂ„ annat Ă€n det de faktiskt ska göra i projektet.

Med det har jag tillĂ„tits fantisera lite och tagit ut svĂ€ngarna. Den fasta punkten i de uttagna svĂ€ngarna Ă€r att IT har en potential i att expandera med en ny “baseline” för den kompetens som ska behövas för att snabbt konstruera större IT projekt och med högre tempo. Ett nytt sorts “alla kan bli programmerare” men dĂ€r innehĂ„llet i uppgifterna Ă€r lĂ„ngt smalare och automatiserat Ă€n vi tĂ€nker idag. Som ett automatiserat produktionsband, men eventuellt utan “kisspausknapp”.


Jonas Nordin | Professional Profile | LinkedIn

View Jonas Nordin’s professional profile on LinkedIn. LinkedIn is the world’s largest business network, helping professionals like Jonas Nordin discover inside connections to recommended job candidates, industry experts, and business partners.