The GNU Foundation has announced that the latest public release of the Bash (version 5.3) command line interpreter is now available, after three long years since the previous stable release.

The new version of Bash includes some interesting new features that will appeal to all types of users. First, let’s talk about what’s new in Bash, and then I’ll show you how you can install the latest release (from source) on your Linux distribution of choice.

What Is Bash?

First, let’s talk about Bash. What exactly is this piece o’ software?

Bash is the most commonly used shell for Linux.

But what is a shell?

In Linux, a shell serves as a command interpreter. Without a shell, you wouldn’t be able to run commands in Linux. When you run a command in Linux, Bash understands those commands and then executes them successfully. Of course, in typical Linux fashion, there are several shells available to choose from. There’s bash, csh, Bourne, KornShell (ksh), T Shell (tcsh), Z Shell, Debian Almquist Shell (dash) and more.

Bash is the default on most Linux distributions because it’s one of the easiest to use, is highly configurable, and includes all the features you would need in a shell. It should be noted that installing new shells for your distribution of choice isn’t always a straightforward experience, but it is very possible.

For example, if you wanted to install the fish shell on Ubuntu, you would first run the command:

sudo apt-get update && sudo apt-get install fish -y

You would then start the fish shell with the command:

fish

You could also make fish the default shell with the Change Shell command:

sudo chsh -s /usr/bin/fish

Your default shell is now fish.

But we’re talking about Bash, so let’s get back to it.

What’s New in Bash?

Version 5.3 adds some significant new features, the most important of which is a new form of command substitution that executes a command in the current shell context. According to the official changelog, “Two forms are implemented: one that reads the command substitution’s output and another that expects to find the result in the REPLY shell variable when the command substitution completes.”

The difference between the two is the addition of the pipe character, like so:

${ command; }
${ | command; }

The first captures the output of the command without forking it, and the second runs the command in the current shell and leaves the results in REPLY.

For those who are unfamiliar with REPLY, it is a default variable that stores the response from the read command when no variable is specified. It allows you to access the input provided by the user without needing to define a separate variable for it.

Then there’s the new GLOBSORT shell variable, which determines how the shell sorts the results of pathname completion.

Other additions include:

• The compgen builtin now has an option to place generated completions into a designated shell variable (vs. standard output).
• A new -E option for the read builtin that uses readline with the default bash completion (including programmable completion).
• A new `-p PATH’ option for the source builtin which forces the user of the PATH argument, as opposed to using $PATH to search for a file.

How To Install Bash 5.3

As of this writing, Bash 5.3 is still in release candidate status, so it’ll be some time before it reaches the standard repositories. If, however, you’re anxious to test out the new features, you can install it from source. Here’s how.

The first step is to download the source. You can either download it using wget with:

wget http://ftp.gnu.org/gnu/bash/bash-5.3.tar.gz

Unpack the file with:

tar -xvzf bash-5.3.tar.gz

Install the necessary dependencies (so you can install from source) with the command:

sudo apt-get install build-essential -y

Change into the bash folder with:

cd bash-5.3

Configure the build with the command:

./configure

Next, you have to compile the code with the command:

make

The above command can take some time to run, so allow it to complete.

Run the install with:

sudo make install

Make sure bash is your default shell with:

sudo chsh -s /usr/local/bin/bash

At this point, you should have Bash 5.3 installed. You can verify the installation with the command:

bash --version

You should see version 5.3.0 listed near the top.

In my opinion, I wouldn’t install 5.3 on a production machine yet because, well, you never know what could happen. If you’re really curious (or want to get up to speed with the new release), I would suggest either installing the latest public release on a spare (nonproduction) machine or in a virtual machine (VM). You certainly don’t want to break your shell on a production machine, as the consequences of that could be devastating (as in, you can no longer run commands … yikes!).

Enjoy that new Bash smell.

The post Bash 5.3 Has Some Big Improvements — Here’s How You Can Test It appeared first on The New Stack.

Today at the AWS Summit in New York City, Swami Sivasubramanian, AWS VP of Agentic AI, provided the day’s keynote on how we’re enabling customers to deliver production-ready AI agents at scale. See below for a roundup of the biggest announcements from the event.

Introducing Amazon Bedrock AgentCore: Securely deploy and operate AI agents at any scale (preview)
Amazon Bedrock AgentCore enables rapid deployment and scaling of AI agents with enterprise-grade security. It provides memory management, identity controls, and tool integration—streamlining development while working with any open-source framework and foundation model.

Announcing Amazon Nova customization in Amazon SageMaker AI
AWS now enables extensive customization of Amazon Nova foundation models through SageMaker AI across all stages of model training. Available as ready-to-use SageMaker recipes, these capabilities allow customers to adapt Nova understanding models across pre-training and post-training, including fine-tuning and alignment recipes to better address business-specific requirements across industries.

AWS Free Tier update: New customers can get started and explore AWS with up to $200 in credits
AWS is enhancing its Free Tier program with up to $200 in credits for new users: $100 upon sign-up and an additional $100 earned by completing activities with services like Amazon EC2, Amazon Bedrock, and AWS Budgets.

TwelveLabs video understanding models are now available in Amazon Bedrock
TwelveLabs video understanding models are now available on Amazon Bedrock and enable customers to search through videos, classify scenes, summarize content, and extract insights with precision and reliability.

Amazon S3 Metadata now supports metadata for all your S3 objects
Amazon S3 Metadata now provides comprehensive visibility into all objects in S3 buckets through live inventory and journal tables, enabling SQL-based analysis of both existing and new objects with automatic updates within an hour of changes.

Introducing Amazon S3 Vectors: First cloud storage with native vector support at scale (preview)
Amazon S3 Vectors is a new cloud object store that provides native support for storing and querying vectors at massive scale, offering up to 90% cost reduction compared to conventional approaches while seamlessly integrating with Amazon Bedrock Knowledge Bases, SageMaker, and OpenSearch for AI applications.

Streamline the path from data to insights with new Amazon SageMaker capabilities
Amazon SageMaker has introduced three new capabilities—Amazon QuickSight integration for dashboard creation, governance, and sharing, Amazon S3 Unstructured Data Integration for cataloging documents and media files, and automatic data onboarding from Lakehouse—that eliminate data silos by unifying structured and unstructured data management, visualization, and governance in a single experience.

Monitor and debug event-driven applications with new Amazon EventBridge logging
Amazon EventBridge now offers enhanced logging capabilities that provide comprehensive event lifecycle tracking, helping users monitor and troubleshoot their event-driven applications with detailed logs that show when events are published, matched against rules, delivered to subscribers, or encounter failures.

Amazon EKS enables ultra scale AI/ML workloads with support for 100K nodes per cluster
Amazon EKS now scales to 100,000 nodes per cluster, enabling massive AI/ML workloads with up to 1.6M AWS Trainium accelerators or 800K NVIDIA GPUs. This allows organizations to efficiently train and run large AI models while maintaining Kubernetes compatibility and existing tooling integration.

At Docker, we always believe in the power of community and collaboration. It reminds me of what Robert Axelrod said in The Evolution of Cooperation: “The key to doing well lies not in overcoming others, but in eliciting their cooperation.” And what better place for Docker Model Runner to foster this cooperation than at Hugging Face, the well-known gathering place for the AI, ML, and data science community. We’re excited to share that developers can use Docker Model Runner as the local inference engine for running models and filtering for Model Runner supported models on Hugging Face!

Of course, Docker Model Runner has supported pulling models directly from HuggingFace repositories for some time now:

docker model pull hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF

Local Inference with Model Runner on Hugging Face

But so far, it has been cumbersome to rummage through the vast collection of models available at HuggingFace and find repositories that work with Docker Model Runner. But not anymore! HuggingFace now supports Docker as a Local Apps provider, so you can select it as the local inference engine to run models. And you don’t even have to configure it in your account; it is already selected as a default Local Apps provider for all users.

Figure 1: Docker Model Runner is a new inference engine available in Hugging Face for running local models.

This makes running a model directly from HuggingFace as easy as visiting a repository page, selecting Docker Model Runner as the Local Apps provider, and executing the provided snippet:

Figure 2: Running models from Hugging Face using Docker Model Runner is now a breeze!

You can even get the list of all models supported by Docker Model Runner (meaning repositories containing models in GGUF format) through a search filter!

Figure 3: Easily discover models supported by Docker Model Runner with a search filter in Hugging Face

We are very happy that HuggingFace is now a first-class source for Docker Model Runner models, making model discovery as routine as pulling a container image. It’s a small change, but one that quietly shortens the distance between research and runnable code.

Conclusion

With Docker Model Runner now directly integrated on Hugging Face, running local inference just got a whole lot more convenient. Developers can filter for compatible models, pull them with a single command, and get the run command directly from the Hugging Face UI using Docker Model Runner as the Local Apps engine. This tighter integration makes model discovery and execution feel as seamless as pulling a container image. 

And coming back to Robert Axelrod and The Evolution of Cooperation, Docker Model Runner has been an open-source project from the very beginning, and we are interested in building it together with the community. So head over to GitHub, check out our repositories, log issues and suggestions, and let’s keep on building the future together.

Learn more

Get an inside look at the design architecture of the Docker Model Runner. 

Explore the story behind our model distribution specification

Read our quickstart guide to Docker Model Runner.

Find documentation for Model Runner.

Visit our new AI solution page

New to Docker? Create an account. 

The technology industry (and beyond) is performative, often offering shallow solutions to deeper problems, shifting the responsibility of change onto those that bear the brunt of the dire consequences the world has created for itself. Want to prevent climate change? Use paper straws instead of holding corporations accountable for their increasing carbon emissions. Want to protect your mental health online? Limit your screen time using app-blocking tools instead of questioning why social media platforms have been deliberately engineered to be addictive. Want to create equal opportunities in the technology industry? Segregate all the women, non-binary, intersex, LGBTQIA+ identifying and trans folk under the banner “Women in Tech” and hope they figure it all out.

At most of the tech companies I have worked, there have been dedicated spaces for “non-men” to get together, including Slack channels, informal meet-ups, and formal discussion panels. “Non-men” in this context has been used to discount “non-heterosexual-cis-men”, and to include women, non-binary, intersex, trans women, trans men and those identifying as LGBTQIA+ all at once. The people spanning the very wide spectrum of this group have probably, and unfortunately, all experienced some kind of discrimination in the workplace, and so it seems sensible for a company to put us all together to share our experiences in a safe and enclosed space, right? It helps, but I'm not so sure it offers a full solution.

The problem with “non-men” spaces

Reports on gender statistics in the technology industry have consistently published figures of around 17-25% of women and non-binary people working in this industry since the early 2000s (this varies by geographical location), and not much has changed over the last 25 years. A recent statistics dump by Spacelift reports that “half of all women who work in tech have left the industry by age 35”, and “none of the biggest US tech companies report women occupying more than a quarter of all technical roles”. Furthermore, “the proportion of undergraduate computer science degrees awarded to women has fallen from 37% in 1985 to about 20% today.”

Why are the statistics not showing much improvement despite countless initiatives surrounding diversity being created, and in particular events, awards, and “safe” spaces for “Women in Tech”? I have a hypothesis: all of these initiatives are operating in closed and exclusive spaces, which is sending a strong signal to the male majority with the power to affect real change that they are not welcome here, and it is not their problem to solve.

A few years ago I was nominated for a “Women in Tech” award by an anonymous party. Not only did I have to confirm my nomination by writing a number of time-consuming essays about my achievements and eligibility for the award, I also couldn’t shake the weird feeling that I was nominated only on the basis that I am a woman. I understand the need to celebrate and lift up “non-men” in tech, and to provide a platform for representation to young and less experienced people entering the industry, but I wasn’t comfortable with demonstrating that if you want to succeed in the tech industry you need special awards and events under the banner of your gender. I did not accept the nomination; I want to win awards that the men are eligible for, too. Because I can, and I have.

These enclosed “non-men” spaces are also not without their specific problems. At one company I worked for, I was asked to participate in a “Senior Women in Tech Panel”, where I was invited to share my experiences and career progression in tech to an audience of women, non-binary, and trans men and women. This event was not moderated well, and descended into a chaotic discussion about how many children each of the panel members had, and negative remarks about the ethnicity and representation of the panel speakers. This may have been a one-off, but at that point I decided to leave the closed Slack channel for “non-men”, and chose not to participate in events of this type for the rest of my career.

What's more, the hard work of ensuring that workplaces are equitable often falls upon the shoulders of underrepresented groups, who are already carrying the burden of discrimination, being overlooked for promotions, and other related issues. A friend reports:

They've started a women's network at work and I am not a member because I think these initiatives are basically just making female members of staff do extra work to fix organisational problems.

The problem with women-coded labels

Imagine if men described themselves as a “male founder” or “HeTO” (a male CTO), or slapped the #MenInTech label on their social media posts when sharing their achievements? The industry’s obsession with women-coded job titles such as “female founder” (which has descended painfully into “girl boss”) that were conjured up to challenge the male default and demonstrate representation in the industry, does not sit comfortably with me.

Feminising job titles, in my opinion, belittles success and achievement in the context of a patriarchal default. Just as “Women in Tech” awards and closed spaces for “non-men” segregate underrepresented groups of people from the spaces where people with power make decisions, women-coded labels inherently classify people on a separate, non-default scale of success and achievement.

It may work for early career folks, but what's next?

After speaking with many women during writing this article, it looks like "Women in Tech" initiatives work well in the context of those entering the industry, but may not be enough to support getting enough underrepresented groups into leadership positions.

A friend of mine shared this with me about her experiences:

I joined tech through Rails Girls. It was a really nice space, it made me feel very safe. It also played into this thing that I knew was happening a lot, which is women thinking that they are not capable of doing something because the world is telling them that they cannot. I really loved getting into tech through these female-focused initiatives. It made me feel like there was a group of people who cared about me joining tech, which is such an important thing if you are looking at an industry that is overloaded with a specific default type of guy.

Another friend had a similar experience, but once she had established her career, things changed:

My perspective also changed with age and expertise gathered. I found a lot of “Women in Tech” programs and initiatives incredibly useful as a newbie. I wouldn’t be where I am without codebar and some early speaking opportunities. But as I gained my voice and my knowledge I found it increasingly patronising.

Another Women in Tech statistics report from AIRPM reports that in 2025, "around 14% of global tech leaders were female in 2023 – up from 8% in 2015", which is promising. Yet, the Spacelift report also finds that “The so-called Big Five tech companies (Alphabet/Google, Apple, Meta/Facebook, Amazon, Microsoft) have never had a female CEO.” And to couple that with another finding from the Spacelift report mentioned above, that “half of all women who work in tech have left the industry by age 35”, it seems that once we get to a certain level of seniority, age, or experience, something isn't working.

Rachel-Lee Nabors wrote about this in 2018 in an article titled A Counterintuitive Way to Increase Diversity in Tech:

I’m mid-career in the web development and design industry. And right now, I’m hearing a lot of the same stories from my fellow mid-career friends from underrepresented groups (women, minorities, LGBTQ, and more): you reach a certain point in your career and you can’t win.

Responses to how you may be feeling right now

I have benefitted from "Women in Tech" initiatives or other "non-men" initiatives in tech. Are you saying these programmes shouldn't exist?

I'm a man and I think DEI is terrible. I'm glad to see a woman agreeing with me at last!

As stated in the disclaimer, this is absolutely not the point of this article. Diversity, equity and inclusion initiatives are absolutely needed in this industry. However, the current landscape has seemingly affected little change in statistics on the whole, especially at leadership levels. I don't think we can continue to segregate ourselves if we want to make progress. Personally, I have worked with some excellent men. And men who have worked with me have said I am also excellent; we have all benefitted from working with each other. To that end, we absolutely need these initiatives: but the men must be invested as well. They must not be excluded from the conversation about how they can help affect change, especially in the context of leadership, given it's mostly the men who hold those positions today.

I am a man and consider myself an ally. Do I need to do something?

I remember speaking on a panel of all women, which was organised by a man. I requested that he not promote the event as an "all-women" event to avoid focussing on gender. He did it anyway. Celebrate underrepresented groups in tech, but do not focus on their gender identity. Celebrate their achievements, use your privilege to lift them up, share their work, make their voice heard: not because they are not men, but because their work is worth celebrating. If you are a man in a leadership position, you can absolutely do more to bring more underrepresented groups into leadership with you. To take another quote from Rachel-Lee's article:

You should be a mouthpiece when someone isn’t heard, an advocate when people are unable to be represented, a shield bearer when folks are under attack, an elevator for the stuck, a signal booster for the unnoticed. Wherever possible, let them do the work, and be heard and understood, and represent themselves. You’re the back up, the path carver, the taker of the brunt of the negative impact, always asking, “How can I help?” Always pushing others up, even over yourself.

We’re focussing on the wrong thing

I think that purely gender-based initiatives are shallow solutions to the deeper problems of male-dominated industries and society as a whole, that focus solely on surface-level optics of inclusion rather than addressing the root cause of inequality and imbalance on a larger scale. So, how do we address the root cause of inequality and imbalance?

We invite those with power to have a seat at the women’s table (and the underrepresented table as a whole): men must be part of the conversation to help affect change, especially at higher seniority levels.

Women, non-binary, intersex and trans folk are in tech, have always been in tech, and will (hopefully) always be in tech. There is a problem, in that these underrepresented groups are often discriminated against and overlooked. The solution? Men in tech can lead the change in how men represent themselves and their gender as a whole in the industry. A bandaid? Groups where "non-men" can feel safe. Both are important, but only one will affect real change.