Vibe coding is great for the App Store economy, but Apple is still wary about its use without safeguards in place. It’s a fine balance that’s going to be hard to maintain.

The concept of vibe coding has risen in tandem with AI chatbots in recent years. Infiltrating many areas of the development process, it has turned the process of making an app into child’s play.

However, while it has its benefits and issues, there are also some areas that Apple is really worried about. Stuff that it is really keen to avoid becoming a serious problem in the future.

It’s something that Apple really wants to sort out now before it becomes a real issue.

What is vibe coding?

Vibe Coding, simply put, is the idea of making an AI chatbot create code for you. At its minimalist level, you tell a chatbot to make something, potentially with some basic specifications or guidance to follow, and it produces the app on its own.

It could be as simple a prompt as “Make me a driving game” or “Create an egg timer that can be adjusted to various durations, and with varying alarm sounds.”

The idea is one that makes it extremely easy for anyone to make an app that they want to use. It takes away the harder elements of learning how to code, how to structure a program, or even knowing how to properly use a development environment like Xcode.

Connect ChatGPT to Xcode and you can ask the AI chatbot to help you make apps.

You could call it a natural but extreme extension of existing code-assistance tools developers already have access to for their work. A developer writing some script can be offered a ready-made element to complete the code, which can save them precious seconds of time.

Vibe coding goes multiple stages further, as you are effectively handing over the majority of the initiative over to an AI. It will follow whatever coding practices it has picked up when scouring the Internet, and produce an app in its own way, fulfilling your prompt.

In some cases, users simply ask for an app to be made and then use the app without examining code or questioning any element. They may give some feedback about a feature, such as asking for changes in font size or color, but the users may not even care about what processes have been undertaken.

The whole concept of telling an AI chatbot what to do goes beyond coding. You can also ask them to create designs and other things in apps, albeit with varying degrees of success.

When it comes to whether vibe coding is a good thing to allow or not, you have to look at it from two angles, as Apple currently does. There’s the case for it being used to make apps, but then there’s the issue of allowing vibe coding to occur beyond the App Store.

Vibe coding as an assistant

One of the areas that is completely fine to Apple is vibe coding outside of the App Store. Making apps to run on a Mac using an AI agent is perfectly fine to Apple.

Indeed, it even encourages it.

For quite some time, you could connect your choice of AI tool to a development environment to make edits to code. With increased integration, the developer environment can be used to the fullest extent by the AI agent, opening the door to full-blown project creation.

This was something that Apple introduced in Xcode 26.3. By enabling more access for AI services, it allowed for the creation of complete apps that followed Apple’s developer guidelines.

This can be a shockingly fast process too. As I found out in February connecting ChatGPT to Xcode, I was able to build a simple Pomodoro timer app and see it working on the Mac screen within two minutes.

A few button presses later, and I saw the exact same app working on a connected iPhone.

Vibe-coding a Pomodoro app within minutes using a simple prompt in Xcode.

This entire process democratizes app development, since you don’t need to know how to use Swift to create an app for your iPhone anymore. If you can describe it well enough to be understood by the AI, and if you don’t care too much about the exact result you get, the app will be usable enough.

Of course, you don’t have to settle for what the AI produces.

AppleInsider readers will be familiar with my development of a game, with the help of ChatGPT in Xcode. Rather than go full-on with vibe coding, I instead took the more careful route of making small and purposeful changes over time.

Every request was considered carefully, with the results checked thoroughly and corrected if needed before moving on. Less vibe coding, more being a project manager in control of a virtual programmer who knows everything about the topic.

It’s entirely possible that someone can vibe code the same game without too much trouble. But at the same time, they won’t know what will happen if something goes wrong and needs to be fixed.

They also won’t necessarily have a result that matches their vision.

The same Pomodoro app on an iPhone, minutes later.

Advertisement

A vibe coder may want to include a feature in their app that they saw elsewhere, and simply ask the AI to do it. But, unless you know exactly how that feature operates and any other underlying elements the other app’s developer took into account, and somehow explain that to the AI perfectly, it won’t be a direct copy.

Sometimes, it’s the execution that actually matters.

With the Xcode integration enhancements, it’s safe to say that Apple is fine with people doing vibe coding in this way. Indeed, it’s something that has helped send a surge of apps through to the App Store review submissions process.

For Apple, vibe coding means more apps in the App Store, and more opportunities to earn from it.

At least, this sort of vibe coding is the version it’s fine with.

Beyond the App Store

The other kind of vibe coding that Apple has to deal with is the version that happens on an iPhone or iPad. This one is a problem because it’s something Apple can’t directly control.

Apple has consistently prevented anyone from being able to compile apps for the iPhone and iPad on the actual hardware. Sure, there’s some coding capability in Swift Playground, but that is a carefully managed experience, and you can’t create a separate standalone app.

This is not a technical limitation of the hardware, as an iPhone or iPad is a computer in its own right and easily capable. But it is a limitation of permissions for a few good reasons.

The App Store Review Guidelines, which app developers must follow, have rules preventing anyone from compiling code. That is, you can’t submit an app that generates code that can then run on the same iPhone or iPad as an app in its own right.

The rules make sense if you consider why the App Store Review Guidelines exist. One of the reasons is to maintain security and privacy, which it does so by checking apps that are submitted to the App Store for potential dangers.

The App Store Review process does accept vibe-coded apps.

For an app capable of coding and compilation, that means it’s possible for apps to be produced beyond the reach of the App Store Review Guidelines. Apple simply cannot check these generated-on-device apps at all.

The worst-case scenario is for someone to make malware on the iPhone or iPad directly, then use it to cause havoc.

With the capabilities of vibe coding as they are, if there are no guardrails in place, a user could create a prompt to make that hazardous software, with no oversight or protections in place.

There is also the argument that Apple would lose potential sales due to users making their own apps. But, in the face of possible malware generation and the potential harm to both the user’s private data and those of other victims, it’s not the one people should be worried about.

Of course, there are ways to enable vibe coding without running afoul of Apple’s guidelines.

For a start, the compilation could happen off-platform. Think of someone making a web app that is hosted on a server elsewhere, but using an iPhone app as a means to design it.

With the more recent development of AI chatbot services where the AI controls a user’s computer elsewhere based on prompts from a mobile app, it’s the same sort of remote-compilation workaround.

Vibe-coding by telling an AI prompt on your iPhone to build something on your Mac isn’t against Apple’s App Store’s rules.

Replit was hit by Apple’s strict rules, but it fixed things and started getting updated again.

Advertisement

A timely example is Apple’s handling of Replit, which saw updates to the app being declined in January. Apple pushed back because it wasn’t happy with users being able to preview AI-built apps on the iPhone, which goes against rules about dynamically executed code.

In May, Replit made a number of changes to appease Apple and allow the app to be updated. The two “worked things out” without there being any real explanation of what changed, but it probably involved changes to that preview mechanism.

Right, but for how much longer?

Apple is, so far, handling the issue of vibe coding pretty well. Really, it’s two issues, and it has gone with the right answer for each one.

Using vibe coding for apps entering the App Store is a completely natural and expected use of the technology. For Apple, the only real downsides are dealing with a greater influx of apps to check and the possibility of the app flood making the market tougher for developers to sell their apps.

These are issues that can be solved by coming up with new ways to process higher numbers of apps, and to aid users in discovering the best apps for their particular needs.

The other side of the equation, the whole business of making apps on an iPhone that aren’t checkable by Apple’s workforce, is also being handled appropriately right now.

Apple is the judge, jury, and executioner when it comes to the App Store’s vibe coding apps.

Apple simply cannot check that a vibe-coded app theoretically running on the iPhone it was created on is actually safe. There are echoes of the whole third-party app storefront arguments when it comes to safety.

Again, I feel Apple is doing the right thing here. It doesn’t want an uncheckable piece of world-destroying malware to be made on an iPhone.

It could be proposed to Apple that it handles apps on an iPhone in the same way as on Mac. A platform that has to deal with apps downloaded from the Internet thanks to a notarization process where Apple checks the app before it becomes downloadable.

There’s more to the issue than that, but macOS proves that it can be done with relative success.

The real problem comes when the major AI players step in and demand that Apple allow for vibe-coded apps to be made and used on the same iPhone.

Despite protestations, Apple will eventually have to decide whether to allow them and deal with the safety and privacy issues, or risk losing out by angering OpenAI, Anthropic, and others.

It’s a problem Apple will have to face and solve at some point down the road. With the rate of change in the AI field, that time could come sooner than anyone thinks.

Jamo isn’t easing back into the market. It’s making noise on multiple fronts ahead of High-End Vienna 2026. After being dropped by Voxx International in 2024 and re-emerging under Cinemaster and Rayleigh Lab, the Danish brand has already returned with the Concert Legacy and Concert Element speaker lines for mid to high-end two-channel and home theater buyers. Now it is moving into a far more crowded Bluetooth speaker category with the new HYG series.

Positioned as a premium lifestyle range built around the Danish concept of hygge, which focuses on comfort, warmth, and connection, the HYG lineup includes three models: Flex, Reflect, and Flow. Each is designed for a different use case, including indoor listening, outdoor spaces, and portable playback. Jamo is not just trying to revive its legacy. It is trying to prove it still belongs in the conversation.

HYG Flex 

The HYG Flex sits at the top of Jamo’s new Bluetooth speaker lineup. It is aimed at listeners who want bigger sound from a portable speaker that still looks like it belongs in a living room, terrace, or outdoor space instead of a dorm-room crime scene.

The cylindrical cabinet is wrapped in soft fabric and designed for easy placement around the home. Inside, dual angled tweeters and a dedicated 5-inch woofer are intended to deliver wider dispersion, stronger bass, and greater scale than the smaller HYG models. Auracast support allows compatible Jamo speakers to connect for multi-speaker playback or stereo pairing, while the built-in ambient base light adds a softer visual touch for evening listening.

HYG Reflect

The HYG Reflect is Jamo’s bedside-focused Bluetooth speaker, designed for listeners who want music, light, and utility in one compact package. The built-in clock makes its intended location rather obvious, and the design leans into calm rather than flashing lights and “party mode” nonsense.

Reflect combines warm ambient lighting, wireless charging, and room-aware audio tuning that adjusts playback based on where the speaker is placed relative to the listener. It is the most lifestyle-focused model in the HYG lineup, built for bedrooms, nightstands, and smaller spaces where a speaker needs to sound good without looking like it escaped from a gaming desk.

The HYG Reflect also includes built-in ambient soundscapes inspired by Scandinavian nature, giving it a relaxation angle beyond basic Bluetooth playback. That makes sense for a bedside speaker, where blasting Slayer at 6:45 a.m. is technically possible but probably not the brand promise.

Two passive radiators and multi-stage tuning are designed to give the Reflect stronger low-frequency output than its compact size might suggest, while the integrated alarm and sleep timer keep the setup simple. The HYG Reflect is not a typical clock radio. It is Jamo’s attempt to make the bedside speaker feel more useful, calmer, and a lot less disposable.

HYG Flow 

HYG Flow is designed primarily for outdoor spaces and road trips. Compact, portable, and IPX7 waterproof-rated, the HYG Flow combines Scandinavian-inspired aesthetics with durable materials and extended battery life. The HYG Flow is designed around portability and simplicity, with a braided handle and nano-coated water-resistant fabric. 

Inside, the HYG Flow uses dual full-range drivers and two passive radiators to deliver balanced stereo sound from a compact portable design. Jamo has tuned it for vocal clarity and longer listening sessions, making it a strong fit for audiobooks, podcasts, radio, and background music that does not punish your ears after 30 minutes.

With up to 27 hours of battery life, Bluetooth 6.0, and Auracast support, the Flow is the most portable member of the HYG lineup and one of the more feature-complete options in Jamo’s new lifestyle range.

JAMO HYG Comparison

The Bottom Line 

Jamo’s HYG line is less about one Bluetooth speaker in three sizes and more about three lifestyle products with different jobs. Flex is the larger portable option for room-filling sound, Reflect is the bedside model with a clock, ambient lighting, wireless charging, alarm, sleep timer, and relaxation soundscapes, while Flow is the grab-and-go speaker with up to 27 hours of playback. That gives the lineup a clearer identity than the usual “small, medium, large” Bluetooth speaker playbook.

The missing pieces matter. There is no Wi-Fi, no app control, and no listed support for LDAC or aptX Lossless, which limits flexibility versus Sonos, JBL, Bose, and other established players. But with pricing from $129 to $279, the HYG series is aimed at listeners who want simple Bluetooth speakers with Scandinavian design, Auracast on select models, and easy placement around the home, garden, bedroom, or bag. Sonos does not need to panic. But Jamo has at least found a sensible way back into the conversation.

Price & Availability

The Jamo HYG Series Bluetooth Speakers will be available globally starting July 2026 through Authorized Dealers at the following prices:

Pro Tip: The Jamo HYG series will be previewed at High End Vienna from June 4 – 7, 2026

Related Reading:

from the buh-bye dept

Senator Bill Cassidy just lost his campaign for reelection to his Senate seat in Louisiana. With that, his career in federal government is likely over. It’s no secret as to why this happened. In early 2021, Cassidy suffered from a spasm of patriotism and voted to convict Donald Trump during his impeachment trial after the latter spurred on an attempted insurrection in the capitol that left several people dead, scores injured, and became the most famous stain on American democracy since the Bill Clinton era. Trump turned his retribution cannons on Cassidy, pumped the primary cycle full of vitriol for Cassidy, and managed to shove him from office.

But here’s the thing: fuck Bill Cassidy.

In the years since Trump’s second impeachment trial, Cassidy did his absolute best to throw as much unrequited love at Donald Trump as he possibly could. The moment the political realities became evident, Cassidy’s principles melted away. He spoke glowingly of Trump in his second term. He touted how well he and Trump work together, even as he acknowledged that Trump hates him. He was a reliable pro-Trump vote on nearly everything.

And he was the most important vote in confirming RFK Jr. to his current role as Secretary of HHS. And given how his vote to confirm Kennedy gave his colleagues cover to do so means that he may be the man most singularly responsible for Kennedy’s appointment other than Donald Trump.

Bill Cassidy is more at fault, because he actually knows better. RFK Jr. is an idiot with a brainworm, whereas Bill Cassidy is a doctor. A real doctor who worked in a charity hospital for the uninsured in Louisiana and, to his own testimony, has seen what happens when children don’t get vaccinated. He knew better, and yet he still provided the deciding vote to confirm Kennedy, almost definitely because he saw it as a way back into Trump’s good graces. Clearly, that didn’t happen.

And while he swore up and down that he had given Kennedy a real good talking-to and got all of the assurances that he would not screw with vaccines or change the CDC’s page stating that vaccines do not cause autism, that did not happen either.

He fucked us all to save his own skin, and ended up bloody and skinless anyway. He claimed it would be okay because of said “assurances” and because of his promise to monitor everything Kennedy did.

Monitor Kennedy he did, perhaps, but it certainly didn’t go beyond that. Save for a few contentious congressional hearings and Cassidy occasionally complaining to local news media about Kennedy’s actions at HHS, the man simply didn’t do anything to try to fix the mess he had a heavy hand in creating. He didn’t sign up to help the impeachment effort against Kennedy. He didn’t call for any new legislation to curb the chaos that is happening at HHS and its child agencies right now. He even had kind words to say about Kennedy’s approach to processed foods in the past few weeks.

Taking a moral stand is not a one-time project. Cassidy’s vote to convict Trump in 2021 was the right vote. Nearly everything he’s done since negates that moral stance, as he engaged in the most pathetic forms of boot-licking in an attempt to save his political career.

It didn’t work. Trump doesn’t work that way. Neither does Kennedy. Once you’re on the enemies hit list, you’re never coming off. The Senate will probably be worse for losing Cassidy generally. The Senate Health Committee certainly will be. And that’s too bad.

But fuck Bill Cassidy for foisting RFK Jr. as HHS boss on this country.

Filed Under: bill cassidy, donald trump, fuck that guy, rfk jr.

It explained that it created the new rules due to Stripe’s policy.

Fitness wearables today are usually marketed toward marathon runners, cyclists, and people already deep into the fitness ecosystem. But for many talented athletes in India, especially those from remote or underrepresented regions, access to proper training tools remains a major challenge. That’s something Garmin now wants to help address through a new initiative in partnership with MyKrida. The company has equipped seven emerging athletes from tribal regions across India with Garmin Forerunner smartwatches to help them access structured performance tracking and training insights.

Garmin Wants to Bring Data-Driven Training to More Athletes

The idea behind the initiative is fairly straightforward. Garmin’s Forerunner smartwatches can track metrics like heart rate, pace, distance, recovery, sleep quality, and training load. For professional athletes, this kind of data is already standard. But for many young athletes in smaller regions, access to these tools can genuinely change how they train. Garmin says the watches are meant to help athletes train smarter and improve consistency through better recovery and performance monitoring rather than simply increasing training intensity.

According to Deepak Raina, Director at AMIT GPS & Navigation LLP:

India has immense untapped athletic potential, particularly in regions where access to structured training tools remains limited. At Garmin, our focus is on enabling athletes with reliable, performance-led technology that brings clarity to how they train, recover, and improve. Through this initiative, we aim to support long-term athletic development and help these athletes compete with greater confidence and consistency.

The on-ground implementation is being managed by MyKrida, which works across grassroots and elite sports development programs in India. The platform focuses heavily on identifying athletes early and connecting them with structured support systems. According to MyKrida founder Shubham Sharma, the collaboration with Garmin helps bring “world-class performance technology directly to these athletes.”

The Dubai-founded global employment and payroll platform reached profitability in early 2025 before taking the round. EBRD Venture Capital led; 500 Global is the new strategic name alongside.\

RemotePass, the global employment, payroll, and spend platform co-founded by Kamal Reggad and Karim Nadi, has raised $17.4m in Series B funding led by EBRD Venture Capital, the company said on Tuesday. 500 Global joins the round as a new strategic investor, alongside returning backers Oraseya Capital, 212 VC, Access Bridge Ventures, and Khwarizmi Ventures.

The round comes 14 months after the company’s $5.5m Series A in March 2024 led by 212 VC, and takes total funding to roughly $23m on the company’s accounting of prior rounds.

The structural detail in the announcement is the profitability claim. RemotePass reached profitability in early 2025 and reinvested the operating-leverage runway in expansion rather than banking it.

Profitable HR-and-payroll-tech companies are unusual in the current vintage; the category is dominated by venture-scale platforms (Deel, Remote, Rippling, Velocity Global) operating on negative unit economics underwritten by repeated late-stage rounds. RemotePass is positioning itself as the discipline-led entrant.

What RemotePass sells is a category that has narrowed in the past two years. The platform handles Employer of Record (EOR) services, contractor management, cross-border payroll and compliance across more than 150 countries, with a fintech layer giving workers USD accounts, global cards and health insurance on top.

The company’s own published numbers put it at 35,000-plus workers across the platform and $800m-plus in cross-border payroll facilitated to date. Named customers include Logitech, Tata Group, InDrive, and Careem.

The product expansion, accelerating the company’s growth, is the fintech layer. In late 2025, RemotePass launched SpendCards, embedding corporate expense cards into the same platform that runs payroll. Expense management has been the operational layer cross-border HR incumbents have struggled with the most; finance teams running distributed workforces have historically had to stitch together payroll providers, expense-card issuers and reimbursement systems. RemotePass’s bet is that the integration of those three is itself the product differentiation.

Chief executive Kamal Reggad framed the round in operational terms: ‘We have the product, the traction, and now the partners to expand properly. Hiring is just the entry point. What companies actually need is a platform that supports their teams end-to-end, including the financial services that make distributed work function.’

The framing is positioning a category convergence as much as a product release. Global employment and embedded fintech, on Reggad’s read, are now the same business; RemotePass is structuring the round around that thesis.

EBRD Venture Capital is the venture arm of the European Bank for Reconstruction and Development, with a mandate across Central and Eastern Europe, the Caucasus, Central Asia and the southern and eastern Mediterranean.

Principal Amine Chabane framed RemotePass as ‘building a leading platform from an emerging market, with the product depth and commercial momentum to compete in Europe and the US’. 500 Global managing partner Amjad Ahmad added that ‘the emerging market depth, embedded fintech layer, and early AI investment create structural advantages that are hard to replicate’.

Deel, RemotePass’s largest competitor, was last reported to be managing in-house payroll teams in over 130 countries and serving 35,000-plus customers. RemotePass’s claim is that the incumbents have underserved cross-border employment cases involving emerging-market entities, banking infrastructure, and compliance complexity.

The MENA-anchored framing is consistent with the wider regional fintech cycle. Dubai has become an increasingly defended fintech hub over the past three years, with the UAE hosting the majority of the region’s most-funded startups and the city itself positioning as a launchpad for emerging-market-founded businesses scaling globally. RemotePass is the workforce-and-fintech variant of that thesis.

The broader labour-market context is tightening. Standard Chartered told investors on Tuesday that the bank would cut more than 15% of its back-office roles by 2030, in part by replacing what its CEO called ‘lower-value human capital’ with AI.

Multiverse’s $70m round earlier this month was framed around the upskilling-versus-replacing trade. RemotePass’s pitch is that it sits inside that intersection.

The company has not disclosed post-money valuation, run-rate revenue, or the geographic split of planned commercial expansion. The next twelve months of new-customer logo announcements will determine whether the Europe-US push lands at scale or whether the company remains an MENA-anchored player with international optionality.

Recently, I wrote about how Apple’s Continuity features keep me locked into the Apple ecosystem because Android and Windows have no answer to Apple’s incredible cross-device seamless integration.

It seems that Android is finally getting started to compete with Apple in this domain. The upcoming Android 17 is bringing a feature that makes switching between your Android devices feel natural. It’s called Continue On, and it lets you start work on one device and seamlessly pick it up on another Android device. 

At launch, the feature focuses on mobile-to-tablet transitions. When you open your tablet, you will see a suggestion in the taskbar for the most recently opened app from your phone. One tap, and you are right back where you started. It’s similar to how Apple’s Handoff feature works. 

How does it work?

Google calls the device you start on the sender and the one you switch to the receiver. The handoff, which is the transition, happens in the background, so you don’t have to do anything complicated. The app just picks up where you left off.

There are a couple of ways this can go, depending on how the developer builds the app. If the app is installed on your tablet, it will deep-link you directly to the activity you were in. Think of opening a Google Docs tab on your phone and having the same document open on your tablet.

What if the app isn’t on your tablet?

If the app is not installed on the tablet, developers can set up a web fallback. So even if the app isn’t installed on the receiving device, it will open the equivalent web experience in your browser instead. Gmail, for example, can hand off from the Android app on your phone to the full Gmail web experience on your tablet, opening the same email thread.

Developers also have the option to skip the app entirely and send users straight to the web version if that is the better experience for a larger screen. Continue On is available in Android 17, and developers can start building support for it now.

It’s highly likely that Google will also add this feature to its upcoming Googlebook laptops, finally providing users with a competitive ecosystem to Apple’s offerings. It’s still early days, but it’s a step in the right direction.

Nintendo plans on getting real weird with it, bringing a new WarioWare-style collection of microgames to mobile platforms just as summer starts.

If you’re a fan of Nintendo’s IP, chances are you’re aware of their propensity for what the publisher calls “microgames.” Mario Party is one famous example, as are games in the WarioWare series.

Microgames are exactly what they sound like.

They’re very, very short games; I’d argue that calling them “minigames” is fairly inaccurate. Each game is designed to be played in mere seconds and is often reaction-time based.

In late May, Nintendo is bringing a new microgame collection to iPhone, iPad, and Android. Dubbed Pictonico!, players use photos they’ve taken for a personalized spin on the genre.

According to Nintendo, you’ll be able to do all the weird things you’d expect. Pluck nose hairs, walk the red carpet, feed your family and friends, strike body building poses, and more.

Nintendo says the game is “free to start,” which means you’ll be able to demo a handful of the games for free. Additional games, of which Nintendo says there are 80, are available as in-app purchases.

If you want to play Pictonico, you can preorder it now ahead of the May 28 launch date. According to the App Store page, your photos are not sent to Nintendo.

In memoriam

We’re not entirely sure how long Pictonico! will last on the App Store. Nintendo has an interesting habit of killing off many of its mobile titles a few years into their lifespan.

Miitomo, Nintendo’s social networking game, made its debut in 2015 and shuttered in very early 2018.

Dr. Mario World suffered a similar fate, launching in 2019 only to be unceremoniously discontinued in 2021.

Perhaps the title with the shortest lifespan of all was Pokemon Rumble Rush, which barely made it past its first birthday before Nintendo pulled the plug.

Of course, sometimes Nintendo doesn’t fully kill off its titles, either.

Mario Kart Tour, for example, launched in 2019. While the game is still available, no major additions have been unveiled since 2023.

Nintendo released Animal Crossing Pocket Camp in 2017, a free-to-play social game for mobile platforms. Seven years after that, the company decided to end its run.

However, instead of simply letting the game go dark, Nintendo created an offline-only version, which is still available.