In this article, we’ll dive into the key features of DeepSeek R1, explore the innovative DeepThinking technology, and discuss how its API integration makes it a game-changer for developers and enterprises.

Use DeepSeek

Go to chat.deepseek.com  for trying deepseek for free. there is no limit for the usage now. also their api is less costly.

What is DeepSeek R1?

DeepSeek R1 is the latest iteration of DeepSeek’s AI language models, designed to tackle even more complex and nuanced tasks. With a focus on reasoning, contextual understanding, and adaptability, R1 builds on the strengths of its predecessors while introducing new capabilities that set it apart from competitors like OpenAI’s GPT-4.

Key features of DeepSeek R1 include:

• DeepThinking: An advanced reasoning engine that enhances problem-solving and decision-making.
• Open-Source Licensing: Released under the MIT License, allowing for free commercial use, modification, and distribution16.
• API Integration: Seamless integration with existing systems for developers and enterprises.
• Scalability: Designed to handle large-scale applications with ease.
• Ethical AI: Continued focus on reducing biases and promoting inclusivity.

DeepThinking: The Brain Behind DeepSeek R1

One of the most exciting innovations in DeepSeek R1 is DeepThinking, a reasoning engine that enhances the model’s ability to reason, analyze, and solve problems. Unlike traditional AI models that rely on pattern recognition, DeepThinking enables R1 to:

1. Break Down Complex Problems: DeepThinking allows the model to dissect intricate problems into smaller, manageable components, making it ideal for tasks like coding, research, and strategic planning14.
2. Simulate Human-Like Reasoning: By incorporating advanced algorithms, DeepThinking enables R1 to simulate human-like reasoning, resulting in more accurate and contextually relevant outputs11.
3. Adapt to New Scenarios: DeepThinking ensures that R1 can adapt to unfamiliar situations, making it a versatile tool for industries like healthcare, finance, and education4.

For example, in a coding scenario, DeepThinking allows R1 to not only generate code but also debug and optimize it, providing a comprehensive solution that goes beyond basic suggestions

Open-Source Licensing and Community Impact

DeepSeek R1 is released under the MIT License, which grants users the freedom to use, modify, and distribute the model for both personal and commercial purposes without restrictions16. This open-source approach fosters collaboration and innovation within the AI community, enabling researchers and developers to build upon DeepSeek R1’s capabilities.

Additionally, DeepSeek has open-sourced six distilled models (ranging from 1.5B to 70B parameters) based on Qwen and Llama architectures. These smaller models retain the reasoning capabilities of R1, making them suitable for resource-constrained environments14.

API Usage: Empowering Developers and Enterprises

DeepSeek R1’s API integration is another standout feature, offering developers and businesses a powerful tool to incorporate AI into their workflows. Here’s how the API can be used:

1. Seamless Integration

The DeepSeek R1 API is designed for easy integration with existing systems, allowing developers to quickly deploy AI capabilities without extensive reconfiguration26.

2. Cost-Effective Pricing

DeepSeek R1’s API is significantly more affordable than competitors like OpenAI, with pricing at $0.55permillioninputtokens\ast \ast and\ast \ast$2.19 per million output tokens28.

3. Real-World Applications

• Healthcare: Use the API to analyze patient data, generate reports, and assist in diagnosis.
• Finance: Integrate R1 for risk assessment, market analysis, and automated trading.
• Education: Create personalized learning experiences and automate administrative tasks.
• Customer Support: Deploy AI-powered chatbots for efficient and accurate customer service

Conclusion

DeepSeek R1 represents a significant leap forward in AI technology, combining the power of DeepThinking with seamless API integration and open-source accessibility. Whether you’re a developer looking to build cutting-edge applications or an enterprise seeking to optimize workflows, DeepSeek R1 offers the tools and capabilities to meet your needs.

With its focus on reasoning, adaptability, and ethical AI, DeepSeek R1 is not just a competitor to GPT-4—it’s a step toward the future of artificial intelligence.

The post DeepSeek R1: Revolutionizing AI with Open Source and DeepThinking appeared first on IB Computing.

How IT Admins Are Fixing the Windows Blue Screen of Death

IT administrators worldwide are in crisis mode today due to a major issue with Windows computers. A recent faulty update from cybersecurity provider CrowdStrike has caused thousands of PCs and servers to crash with the notorious Windows Blue Screen of Death (BSOD) error. Although CrowdStrike has resolved the update issue, countless systems remain offline, affecting banks, airlines, supermarkets, and TV broadcasters.

The Fix: Tackling the Windows Blue Screen of Death

Windows Blue Screen of Death Fix  for many affected systems isn’t straightforward. IT admins are utilizing a suggested workaround from CrowdStrike, which involves booting Windows into Safe Mode and manually deleting the problematic system file to initiate the Windows Blue Screen of Death fix:

1. Boot Windows into Safe Mode or the Windows Recovery Environment.
2. Navigate to the C:\Windows\System32\drivers\CrowdStrike directory.
3. Locate and delete the file matching “C-00000291*.sys”.
4. Reboot the host.

These steps help boot Windows into an environment where third-party drivers, such as CrowdStrike’s kernel-level driver, aren’t loaded. Admins must physically access machines to delete the faulty driver, complicating matters in environments with disk encryption like BitLocker or limited admin rights.

Challenges and Solutions

For many environments, the workaround is complicated by technical constraints, pushing some IT admins to wait for CrowdStrike’s fix. This entails repeated reboots in hopes that the update gets applied before the machine crashes again. Surprisingly, this “turn-it-off-and-on-again” method has had some success, with machines coming back online after multiple reboots. The delay can be attributed to CrowdStrike’s update servers being overwhelmed by an influx of requests from millions of machines.

Businesses using virtual desktops might recover faster by restoring systems to a pre-update state. However, in cases where reboots fail, booting into Safe Mode remains the most reliable option to implement a Windows Blue Screen of Death fix.

CrowdStrike’s Response

This crisis won’t be resolved in just a few hours. “It could take some time for systems that won’t automatically recover, but it is our mission to ensure every customer is fully operational,” stated CrowdStrike CEO George Kurtz in an interview with NBC News. Kurtz apologized for the widespread disruptions caused by the faulty update, promising a thorough investigation into how such an issue could affect so many machines globally.

The post IT Admins Battle Windows Blue Screen of Death After Faulty Update appeared first on IB Computing.

Google’s Goo.gl Link Shortener Expire on Aug 2025

Hold onto your hyperlinks, folks!  Google’s Goo.gl Link Shortener Expire soon. it  is officially nearing the end of its digital road. After ceasing the creation of new goo.gl URLs in March 2019, Google has now set the definitive expiration date for these links: August 25, 2025.

The End of an Era

The sun began to set on goo.gl back in 2018 when Google first announced its shutdown for consumers and developers. A year later, the company halted the creation, analytics, and management of new links for all users. Fast forward to 2025, and it’s “404 Not Found” for any goo.gl URL still roaming the web.

Transition Period Ahead

Starting August 23, 2024, users clicking on a goo.gl link will encounter an interstitial page saying, “This link will no longer work in the near future.” Fear not, you can hit “Continue” to proceed, with an option to “Don’t show this again” for future clicks.

Developers, Take Note!

Google has a special note for developers: beware of potential disruptions! The interstitial page might throw a wrench in your redirect flows and social metadata displays. You might want to start shifting your goo.gl links ASAP. As an interim fix, adding the query parameter “si=1” should suppress the interstitial page for the time being.

Here’s the official word from Google:

“For example, if you are using other 302 redirects, the interstitial page may prevent the redirect flow from completing correctly. If you’ve embedded social metadata in your destination page, the interstitial page will likely cause these to no longer show up where the initial link is displayed. For this reason, we advise transitioning these links as soon as possible.”

The Next Step: Firebase Dynamic Links

Back in 2018, Google nudged developers to shift to Firebase Dynamic Links, designed to detect the user’s platform and direct them to either the web or an app. Oh, the irony! Google eventually pulled the plug on that service for developers too.

Time to Update Your URLs

So, mark your calendars for August 25, 2025 Google’s Goo.gl Link Shortener Expire. Whether you’re a developer or just a goo.gl enthusiast, it’s time to update those shortened links before they become digital relics.

The post Google’s Goo.gl Link Shortener Expire in Aug 2025 appeared first on IB Computing.

After years of anticipation, Wikipedia introduces dark mode, now available on select wikis for both mobile and desktop in reading and editing modes. This feature, one of the most requested by the community, aims to enhance accessibility and reduce eye strain by providing a low-contrast environment.

Why Dark Mode Matters

Dark mode hasn’t just been a nice-to-have; it’s been a necessity. Since at least 2019, editors and readers have pushed for this feature, with 20% signaling their dark mode preference at the OS level. Though research is mixed, many agree dark mode lessens eye strain and helps those with certain medical conditions.

Overcoming Challenges

Creating dark mode for Wikipedia involved overcoming significant technical hurdles. The new Vector 2022 skin was pivotal, and the collaborative effort between Wikimedia Foundation engineers and volunteer editors was essential. Articles like “International Orange” demonstrated the necessity of precision to avoid misinformation caused by altered color presentations.

The Accessibility Opportunity

Introducing dark mode highlighted pre-existing accessibility issues in many articles. For instance, the Wrexham A.F.C. article’s inaccessible red headers were revamped for better readability. This project has spotlighted the importance of wise color usage and overall content improvement.

How to Activate Dark Mode

Dark mode is currently available for logged-in users across all Wikipedias and gradually rolling out to non-logged-in users on wikis with minimal color issues or active technical editor involvement. Supported skins include Minerva (mobile) and Vector 2022 (desktop). Editors proficient in CSS and templates are encouraged to help expand dark mode availability.

A Collaborative Achievement

The rollout has been a collective effort, featuring invaluable contributions from volunteer editors and various Wikimedia teams. This milestone wouldn’t have been possible without their dedication and collaboration.

Enjoy the Experience!

As dark mode continues to expand, users are invited to enjoy a more comfortable reading experience on Wikipedia. For more info and to help improve accessibility, volunteer your expertise on CSS and templates.

Stay tuned as Wikipedia keeps evolving for the better – one dark mode at a time!

The post Wikipedia Launches Long-Awaited Dark Mode Feature appeared first on IB Computing.

Natively supported apps

• Discord
• Gitter
• IRC
• Keybase
• Matrix
• Mattermost 4.x, 5.x
• Microsoft Teams
• Nextcloud Talk
• Rocket.chat
• Slack
• Ssh-chat
• Steam
• Telegram
• Twitch
• WhatsApp
• XMPP
• Zulip

The whatsapp linking part is based on go-whatsapp which is a Go version of  whatsapp-web-reveng.

I think that’s enough for the intro, let’s get back to the main point

Install and Run

Download the latest version from here
select the required package. im using linux for this tutorial so i select matterbridge-1.18.3-linux-64bit (this version will change)
then move to the Configuration part

Step 1

Create an empty matterbridge.toml file inside a folder contains downloaded matterbridge binary file.

Step 2

Copy the following contents to the newly created matterbridge.toml file:

[telegram.mytelegram]
#See https://core.telegram.org/bots#6-botfather 
#and https://www.linkedin.com/pulse/telegram-bots-beginners-marco-frau
Token="Yourtokenhere"
RemoteNickFormat="({PROTOCOL}) <{NICK}> "
MessageFormat="HTMLNick"

[whatsapp.mywhatsapp]
# Number you will use as a relay bot. Tip: Get some disposable sim card, don't rely on 
# your own number.
Number="+48111222333"

# First time that you login you will need to scan QR code, then credentials will be saved in 
# a session file. 
# If you won't set SessionFile then you will need to scan QR code on every restart.
# optional (by default the session is stored only in memory, till restarting matterbridge)
SessionFile="session-48111222333.gob"

# If your terminal is white we need to invert QR code in order for it to be scanned properly
# optional (default false)
QrOnWhiteTerminal=false

# Messages will be seen by other WhatsApp contacts as coming from the bridge. 
# Original nick will be part of the message.
RemoteNickFormat="[{PROTOCOL}] @{NICK}: "

Step 3

Change the Token and Server keys above to yours. For more details you may watch this video.

In Telegram, go to botfather and create a bot and paste the token as Token="Yourtokenhere"
Enter your whatsapp number as Number="+48111222333"

Step 4

Add the gateway configuration to your config file.

[[gateway]]
name="gateway1"
enable=true

[[gateway.inout]]
account="whatsapp.mywhatsapp"
channel="[email protected]"

[[gateway.inout]]
account="telegram.mytelegram"
channel="-xxxxx"

Step 5

Now your configuration file should look like this:

[telegram.mytelegram]
#See https://core.telegram.org/bots#6-botfather 
#and https://www.linkedin.com/pulse/telegram-bots-beginners-marco-frau
Token="1279548291:AAFg3s7F-JQlQQlucSBf3PGwyOss2Vog0J4"
RemoteNickFormat="({PROTOCOL}) <{NICK}> "
MessageFormat="HTMLNick"

[whatsapp.mywhatsapp]
# Number you will use as a relay bot. Tip: Get some disposable sim card, don't rely on 
# your own number.
Number="+48111222333"

# First time that you login you will need to scan QR code, then credentials will be saved in 
# a session file. 
# If you won't set SessionFile then you will need to scan QR code on every restart.
# optional (by default the session is stored only in memory, till restarting matterbridge)
SessionFile="session-48111222333.gob"

# If your terminal is white we need to invert QR code in order for it to be scanned properly
# optional (default false)
QrOnWhiteTerminal=false

# Messages will be seen by other WhatsApp contacts as coming from the bridge. 
# Original nick will be part of the message.
RemoteNickFormat="[{PROTOCOL}] @{NICK}: "

[[gateway]]
name="gateway1"
enable=true

[[gateway.inout]]
account="whatsapp.mywhatsapp"
channel="[email protected]"

[[gateway.inout]]
account="telegram.mytelegram"
channel="-xxxx"

Step 6

Find channel ID for Telegram and WhatsApp. Finding Telegram group ID is relatively easy.  You can use the Telegram bot API to get the group ID. But I’m taking a simple shortcut here.  Adding another bot to retrieve group ID, which is @tchncs_bot. This is a bot that bridges Matrix and Telegram groups. Just add the bot and sending the message /id (including the forward slash, which makes the message a command) will return the group ID.

Unlike in Telegram’s case, finding channel name in WhatsApp isn’t that easy. I couldn’t find any easier method from matterbridge documentation as well. So enter your WhatsApp group name there and run the bridge (Warning: you will definitely get an error!)

Step 7

Run it with the following command (change the version number if you are using later version)
./matterbridge-1.18.3-linux-64bit -conf matterbridge.toml

Now you can see a QR code in terminal. If it is too big that you weren’t able to scan it, then zoom in the terminal and try again. Then scan the QR code with WhatsApp web option from your mobile phone. Now your WhatsApp web is connected with the matterbridge. Next step is to identify the channel name for WhatsApp. By checking the terminal output you can see your group name and a number that ends with @g.us. Copy that and paste it in the channel part of WhatsApp and run again. Now your bridge is ready and it will start working. You can also share image, files and audios using this bridge. If you want to bridge multiple groups the gateway part can be changed to

[[gateway]]
name="gateway1"
enable=true

[[gateway.inout]]
account="whatsapp.mywhatsapp"
channel="[email protected]"

[[gateway.inout]]
account="telegram.mytelegram"
channel="-xxxx"

[[gateway]]
name="gateway2"
enable=true

[[gateway.inout]]
account="whatsapp.mywhatsapp"
channel="[email protected]"

[[gateway.inout]]
account="telegram.mytelegram"
channel="-xxxx"

Drawbacks

This brdige reies on WhatsApp web. So if you have to make this solution work, you should be online on both mobile as well as on WhatsApp web all the time. Other solution you could try is to run any emulator on server. Also, the matterbridge should be running all the time. So you have to choose a server for that. Cloudcone provides low cost servers with good quality. You can use this link to using cloudcone.

The post Setting up WhatsApp Telegram Bridge Using Matterbridge 2020 appeared first on IB Computing.

We have to change the font globally because Telegram doesn’t have the option to change font.

Change font in Telegram desktop application in Gnu-Linux OS

Create a file at the location ~/.config/fontconfig/fonts.conf

and add the following content to it. It sets default sans-serif font to Meera and serif font to Rachana. change the font name as your wish

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE fontconfig SYSTEM "fonts.dtd">
<!— —>
<fontconfig>
<!— Generic name aliasing —>
<alias>
<family>sans-serif</family>
<prefer>
<family>Meera</family>
</prefer>
</alias>
<alias>
<family>serif</family>
<prefer>
<family>Rachana</family>
</prefer>
</alias>
<alias>
<family>monospace</family>
<prefer>
<family>Liberation Mono</family>
</prefer>
</alias>
</fontconfig>

then run

 fc-cache -v.

and restart Telgram Desktop application. this will also change your fonts in all applications like browser.

In – Windows Operating System

Open Registy editor

press Win+R and type regedit

go to the location given below

HKEY_LOCAL_MACHINE
SOFTWARE
Microsoft
Windows NT
CurrentVersion
FontSubstitutes

Change Value of “MS Shell Dlg 2” to the font name

I don’t have a Mac system, if anyone of you know about it add it in comments. This will change the fonts in Telegram Desktop application.

The post How to change font in telegram desktop in Linux and Windows appeared first on IB Computing.

Reveal js is a JavaScript package to create web slides.  official website itself a presentation demo.  it also provides a free online editor to create slides . but it needs an account to start. just go to slide.com for making one.  lets see how we can create a web slide using a markdown file.

Build Presentations with Reveal.js and markdown

Before we start we have to install the following packages into the system.

execute these commands to install the packages in Ubuntu. Other operating systems have there own package manager commands.

git : sudo apt-get install git
pandoc : sudo apt-get install pandoc

step 1 : Create a Markdown File with headings and list items. an example here

django-intro.md

% Django Introduction
% Mujeeb Rahman K
% February 09, 2020

# Django

## Why Django
* Django is a Web framework written in Python
* Don't reinvent the wheel.
* 216 K packages in python package index

step 2 : clone or download Revealjs into the same folder which contains the above markdown file

git clone https://github.com/hakimel/reveal.js.git

Step 3 : convert markdown (.md) file to html file using pandoc utility

pandoc -t revealjs -s -o djangoslide.html django-intro.md -V revealjs-url=./reveal.js

This will convert md file to html and use reveal js from our current folder. if you don’t cloned reveal js you can use direct url like below.

pandoc -t revealjs -s -o djangoslide.html django-intro.md -V revealjs-url=https://revealjs.com

now we have got a djangoslide.html file. just open that file in any browser and see the magic..!! your web slide is ready.  remember that, if you didn’t download or clone reveal js in same folder the html wont render correctly.

The is a very quick and simple way to create a presentation because it only need very simple utilities like git, pandoc and a simple markdown file. Markdown can be created using a very simple text editor. We have very nice presentation with all the effects. It is html so simply works everywhere.

The post Create A Simple Presentation using Reveal js and Markdown appeared first on IB Computing.

Ubuntu 19.04 is the next stable Ubuntu release and is due for release in April 2019 as expected. It’ll have support for at least nine months since the original stable release, which will be until July 2020.

Ubuntu just recently released its latest stable version 18.10 named Cosmic Cuttlefish last month. Since the Ubuntu 19.04 daily test build ISO is based on this don’t look for any exciting new and/or fancy features or improvements. You can now download the first official Ubuntu 19.04 Disco Dingo daily build ISO image directly from the Canonical website.

Download Ubuntu 19.04 Disco Dingo Daily Build ISO

It should be kept in mind that this is a very early development version which may have bugs and undetected issues, so it is never ever recommended to install on production machines. As mentioned earlier it is targeted at early testers for testing purpose.

The Ubuntu 19.04 Disco Dingo beta release is expected in March 28, 2019. If you don’t want to waste time for just testing then you can wait until the beta release. Or if you immediately want a new and stable operating system then you can of course download the recently released Ubuntu 18.10 Cosmic Cuttlefish. It comes with a new default GNOME theme called Yaru and also features Suru icon pack. With Linux kernel v4.18 and GNOME v3.30 there’s also boot time as well as performance improvements.

Download Ubuntu 18.10 Cosmic Cuttlefish

The post Download Ubuntu 19.04 Disco Dingo Daily Build ISO appeared first on IB Computing.

lspci -nn | grep -i network

If you are having the Broadcom BCM43142 WiFi device then you will get an output somewhat like follows:

0d:00.0 Network controller [0280]: Broadcom Limited BCM43142 802.11b/g/n [14e4:4365] (rev 01)

If you get the above output then we are good to continue.

How to install WiFi driver for Broadcom BCM43142 in Debian and Debian-based distros like Ubuntu

sudo apt-get update
sudo apt-get install broadcom-sta-dkms

The advantage of using a dkms driver is that you don’t need to manually update the driver everytime you upgrade your kernel!

Dynamic Kernel Module Support (DKMS) is a program/framework that enables generating Linux kernel modules whose sources generally reside outside the kernel source tree. The concept is to have DKMS modules automatically rebuilt when a new kernel is installed.
– Wikipedia

How to install WiFi driver for Broadcom BCM43142 WiFi device in Fedora

sudo dnf update
sudo dnf install https://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm https://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm
sudo dnf update && sudo dnf install kmod-wl

How to install WiFi driver for Broadcom BCM43142 WiFi device in Arch Linux

sudo pacman -S broadcom-wl-dkms

The post How to Install WiFi driver for Broadcom BCM43142 WiFi device in GNU/Linux Distros appeared first on IB Computing.

Whats New?

KDE Plasma 5.14 comes with lot of new changes, but my favorite is the Display Configuration widget for screen management. It was always a mess when it comes to doing presentations, pressing buttons, not pressing buttons, mirroring on, mirroring off, it was damn hard to properly set up a projector. There are applications like arandr, but now there is a KDE widget to control that and it is very very very handy. KDE plasma 5.14 comes with many other new features too, here are some of them.

* Audio Volume widget has a built in speaker test feature.
* Switching primary monitor when plugging and unplugging monitors is smoother.
* Modifications on Lock Screen, now easier for switching user.
* Plasma vault supports exporting encrypted files.
* System monitor has a Tools menu where you can launch other utilities.

* Kickoff application menu switches tab instantly on hover.
* Warns about logged in users while logging out.

* Global menu now supports GTK apps too.

New in Plasma Discover

* With fwupd support, now Discover supports firmware updates.
* Snap Channel support.
* Now apps can be sorted by release date.
* Package dependencies can be listed.
* Improved Flatpak support.
* Many interface improvements.
* More stable, multiple bug fixes.

Kwin and Wayland

* Fixed copy paste between GTK and non GTK applications.
* Improved KWin for sticker animations.
* Multiples effects have rewritten for smoother performance.
* Improved glide effect, adding new scale effect.

And many bugfixes too. For further information check the KDE’s release announcement.

You can try the latest KDE Plasma using KDE neon. You can download latest KDE neon from here.

Download KDE neon

The post KDE Plasma 5.14 Released appeared first on IB Computing.